INSIGHT4G Software Manual

INSIGHT 4G™
GLOBAL IMAGE,
ACQUISITION, ANALYSIS,
& DISPLAY SOFTWARE
USER’S GUIDE
P/N 6004904, REVISION C

MARCH 2015
Manual History
This is a manual history of the Insight 4G™ software manual

(Part Number 6004904).
Revision Date
A June 2011
B June 2014
C March 2015
This manual is used with Insight 4G (all versions) software.
ii
Warranty
Part Number 6004904 / Revision C / March 2015

Copyright © TSI Incorporated / 2011-2015 / All rights reserved.
Address TSI Incorporated / 500 Cardigan Road / Shoreview, MN 55126 / USA
Phone No. 1-800-874-2811 (USA) or (651) 490-2811
Fax No. (651) 490-3824
E-mail Address [email protected]
Limitation of Warranty Seller warrants the goods, excluding software, sold hereunder, under normal use and service
and Liability as described in the operator's manual, to be free from defects in workmanship and material for
(effective April 2014) 12 months, or if less, the length of time specified in the operator's manual, from the date of
shipment to the customer. This warranty period is inclusive of any statutory warranty. This
limited warranty is subject to the following exclusions and exceptions:
a. Hot-wire or hot-film sensors used with research anemometers, and certain other
components when indicated in specifications, are warranted for 90 days from the date of
shipment;
b. Pumps are warranted for hours of operation as set forth in product or operator’s manuals;
c. Parts repaired or replaced as a result of repair services are warranted to be free from
defects in workmanship and material, under normal use, for 90 days from the date of
shipment;
d. Seller does not provide any warranty on finished goods manufactured by others or on any
fuses, batteries or other consumable materials. Only the original manufacturer's warranty
applies;
e. This warranty does not cover calibration requirements, and seller warrants only that the
instrument or product is properly calibrated at the time of its manufacture. Instruments
returned for calibration are not covered by this warranty;
f. This warranty is VOID if the instrument is opened by anyone other than a factory
authorized service center with the one exception where requirements set forth in the
manual allow an operator to replace consumables or perform recommended cleaning;
g. This warranty is VOID if the product has been misused, neglected, subjected to accidental
or intentional damage, or is not properly installed, maintained, or cleaned according to the
requirements of the manual. Unless specifically authorized in a separate writing by Seller,
Seller makes no warranty with respect to, and shall have no liability in connection with,
goods which are incorporated into other products or equipment, or which are modified by
any person other than Seller.
The foregoing is IN LIEU OF all other warranties and is subject to the LIMITATIONS stated
herein. NO OTHER EXPRESS OR IMPLIED WARRANTY OF FITNESS FOR PARTICULAR
PURPOSE OR MERCHANTABILITY IS MADE. WITH RESPECT TO SELLER’S BREACH
OF THE IMPLIED WARRANTY AGAINST INFRINGEMENT, SAID WARRANTY IS LIMITED
TO CLAIMS OF DIRECT INFRINGEMENT AND EXCLUDES CLAIMS OF CONTRIBUTORY
OR INDUCED INFRINGEMENTS. BUYER’S EXCLUSIVE REMEDY SHALL BE THE
RETURN OF THE PURCHASE PRICE DISCOUNTED FOR REASONABLE WEAR AND
TEAR OR AT SELLER’S OPTION REPLACEMENT OF THE GOODS WITH NON-
INFRINGING GOODS.
TO THE EXTENT PERMITTED BY LAW, THE EXCLUSIVE REMEDY OF THE USER OR
BUYER, AND THE LIMIT OF SELLER'S LIABILITY FOR ANY AND ALL LOSSES, INJURIES,
OR DAMAGES CONCERNING THE GOODS (INCLUDING CLAIMS BASED ON
CONTRACT, NEGLIGENCE, TORT, STRICT LIABILITY OR OTHERWISE) SHALL BE THE
RETURN OF GOODS TO SELLER AND THE REFUND OF THE PURCHASE PRICE, OR,
AT THE OPTION OF SELLER, THE REPAIR OR REPLACEMENT OF THE GOODS. IN THE
CASE OF SOFTWARE, SELLER WILL REPAIR OR REPLACE DEFECTIVE SOFTWARE
OR IF UNABLE TO DO SO, WILL REFUND THE PURCHASE PRICE OF THE SOFTWARE.
IN NO EVENT SHALL SELLER BE LIABLE FOR LOST PROFITS, BUSINESS
INTERRUPTION, OR ANY SPECIAL, INDIRECT, CONSEQUENTIAL OR INCIDENTAL
DAMAGES. SELLER SHALL NOT BE RESPONSIBLE FOR INSTALLATION, DISMANTLING
OR REINSTALLATION COSTS OR CHARGES. No Action, regardless of form, may be
brought against Seller more than 12 months after a cause of action has accrued. The goods
iii
returned under warranty to Seller's factory shall be at Buyer's risk of loss, and will be returned,
if at all, at Seller's risk of loss.
Buyer and all users are deemed to have accepted this LIMITATION OF WARRANTY AND
LIABILITY, which contains the complete and exclusive limited warranty of Seller. This
LIMITATION OF WARRANTY AND LIABILITY may not be amended, modified or its terms
waived, except by writing signed by an Officer of Seller.
Software License This is a legal agreement between you, the end user, and TSI Incorporated. BY INSTALLING
THE SOFTWARE, YOU ARE AGREEING TO BE BOUND BY THE TERMS OF THIS
(effective March 1999) AGREEMENT. IF YOU DO NOT AGREE TO THE TERMS OF THIS AGREEMENT,
PROMPTLY RETURN THE UNOPENED PACKAGE AND THE ACCOMPANYING ITEMS
(including written materials and binders or other containers) to TSI for a full refund.
1. GRANT OF LICENSE. TSI grants to you the right to use one copy of the enclosed TSI
software program (the “SOFTWARE”), on a single computer. You may not network the
SOFTWARE or otherwise use it on more than one computer or computer terminal at the same
time.
2. COPYRIGHT. The SOFTWARE is owned by TSI and is protected by United States copyright
laws and international treaty provisions. Therefore, you must treat the SOFTWARE like any
other copyrighted material (e.g., a book or musical recording) except that you may either (a)
make one copy of the SOFTWARE solely for backup or archival purposes, or (b) transfer the
SOFTWARE to a single hard disk provided you keep the original solely for backup or archival
purposes.
3. OTHER RESTRICTIONS. You may not rent or lease the SOFTWARE, but you may transfer
the SOFTWARE and accompanying written material on a permanent basis, provided you retain
no copies and the recipient agrees to the terms of this Agreement. You may not reverse-
engineer, decompile, or disassemble the SOFTWARE.
4. DUAL MEDIA SOFTWARE. If the SOFTWARE package contains multiple types of media,
then you may use only the media appropriate for your single-user computer. You may not use
the other media on another computer or loan, rent, lease, or transfer them to another user
except as part of the permanent transfer (as provided above) of all SOFTWARE and written
material.
5. U.S. GOVERNMENT RESTRICTED RIGHTS. The SOFTWARE and documentation are
provided with RESTRICTED RIGHTS. Use, duplication, or disclosure by the Government is
subject to the restrictions set forth in the “Rights in Technical Data and Computer Software”
Clause at 252.227-7013 and the “Commercial Computer Software - Restricted Rights” clause at
52.227-19.
6. LIMITED WARRANTY. TSI warrants that the SOFTWARE will perform substantially in
accordance with the accompanying written materials for a period of ninety (90) days from the
date of receipt.
7. CUSTOMER REMEDIES. TSI’s entire liability and your exclusive remedy shall be, at TSI’s
option, either (a) return of the price paid or (b) repair or replacement of the SOFTWARE that
does not meet this Limited Warranty and which is returned to TSI with proof of payment. This
Limited Warranty is void if failure of the SOFTWARE has resulted from accident, abuse, or
misapplication. Any replacement SOFTWARE will be warranted for the remainder of the
original warranty period or thirty (30) days, whichever is longer.
8. NO OTHER WARRANTIES. TSI disclaims all other warranties, either express or implied,
including, but not limited to implied warranties of merchantability and fitness for a particular
purpose, with regard to the SOFTWARE and the accompanying written materials.
9. NO LIABILTY FOR CONSEQUENTIAL DAMAGES. In no event shall TSI be liable for any
damages whatsoever (including, without limitation, special, incidental, consequential or indirect
damages for personal injury, loss of business profits, business interruption, loss of information
or any other pecuniary loss) arising out of the use of, or inability to use, this SOFTWARE.
Service Policy Knowing that inoperative or defective instruments are as detrimental to TSI as they are to our
customers, our service policy is designed to give prompt attention to any problems. If any mal-
function is discovered, please contact your nearest sales office or representative, or call TSI at
1-800-874-2811 (USA) or (651) 490-2811.
Trademarks TSI logo, Insight, Insight 3G, Insight 4G, and PowerView are trademarks of TSI Incorporated.
TSI is a registered trademark of TSI Incorporated.
Microsoft, Windows, and Excel are registered trademarks of Microsoft Corporation in the United
States and other countries.
FireWire is a registered trademark of Apple Inc., registered in the U.S. and other countries.
Photron is a registered trademark of Photron Technologies Ltd.
Tecplot is a registered trademark of Tecplot, Inc., Bellevue, WA, USA.
MATLAB is a registered trademark of The MathWorks, Inc.
iv Insight 4G Global Image Acquisition, Analysis and Display Software

Contents
Manual History .......................................................................................... ii

Warranty ................................................................................................... iii
Contents .................................................................................................... v
About This Manual .................................................................................. ix
Purpose ................................................................................................ix
Related Product Literature ...................................................................ix
Submitting Comments ..........................................................................ix
CHAPTER 1 Introduction ...................................................................... 1-1
CHAPTER 2 Getting Started ................................................................. 2-1
Unpacking and Checking Package Contents ................................... 2-1
Hardware and Software Requirements ............................................. 2-2
Installing Insight Software ................................................................. 2-2
Starting Insight Software ................................................................... 2-2
CHAPTER 3 Getting Acquainted with Insight Software .................... 3-1
The Insight Desktop .......................................................................... 3-1
Insight Tool Bar ................................................................................. 3-3
Control Panel .................................................................................... 3-4
Display Panel .................................................................................... 3-4
Status Bar ......................................................................................... 3-4
CHAPTER 4 Setting up Insight Software ............................................ 4-1
Specifying the Experiment Folder ..................................................... 4-1
Specifying Hardware Components in Your Experiment.................... 4-2
Using Dual Monitors .......................................................................... 4-3
CHAPTER 5 Creating and Managing Experiment and Data Files ..... 5-1
Creating Experiment and Run Folders ............................................. 5-1
Managing Insight Files ...................................................................... 5-2
CHAPTER 6 Viewing, Enhancing, and Displaying Image Files......... 6-1
Display Panel Features ..................................................................... 6-2
CHAPTER 7 Capturing Images ............................................................. 7-1
Capturing Images: An Overview ....................................................... 7-1
Setting Hardware Component Parameters ....................................... 7-2
Enabling and Setting Traverse.......................................................... 7-8
Setting Timing Parameters ............................................................... 7-8
Selecting the Exposure Mode ......................................................... 7-11
Selecting the Capture Mode ........................................................... 7-11
Selecting Laser Power and Pulse Energy ...................................... 7-13
Capturing Calibration Images ......................................................... 7-14
Starting and Stopping Image Captures ........................................... 7-15
Saving, Loading and Deleting Image Capture Settings .................. 7-15
Saving Image Captures .................................................................. 7-16
CHAPTER 8 Processing Images .......................................................... 8-1
Processing Tab ................................................................................. 8-1
v
Defining a Static Processing Mask ................................................... 8-3
Defining a Dynamic Processing Mask .............................................. 8-5
Apply a Processing Mask ................................................................. 8-9
Deleting a Processing Mask ............................................................. 8-9
Using a Region of Interest (ROI) ...................................................... 8-9
Using Image Preprocessing............................................................ 8-11
Image Preprocessors...................................................................... 8-12
Size Shape Analysis ....................................................................... 8-21
CHAPTER 9 2-D Calibration and Perspective Calibration................ 9-1
2-D Calibration Process .................................................................... 9-1
Perspective Calibration Process ....................................................... 9-4
Perspective Calibration Overview ..................................................... 9-4
Step A: Overlapping the Camera Fields of View with the
Calibration Target .......................................................................... 9-8
Step B: Setting Scheimpflüg Angle and Focus with Particle
Images, Record Flow Images for AutoMapping ............................ 9-9
Step C: Recording Calibration Images ........................................... 9-10
Step D: Processing the Calibration Images and Creating a
New Calibration File .................................................................... 9-12
Step E: Stereo AutoMapping .......................................................... 9-32
CHAPTER 10 Processing and Viewing PIV Images ......................... 10-1
Rules-of-Thumb for PIV Analysis ................................................... 10-2
PIV Cross-Correlation Processing ................................................ 10-11
Setting up PIV Processor.............................................................. 10-13
PIV Plugin Engines ....................................................................... 10-15
Double Correlation Analysis ......................................................... 10-24
PIV Process Monitor ..................................................................... 10-25
Particle Tracking Analysis (Super-Resolution Particle
Velocimetry)............................................................................... 10-26
PIV Vector Validation .................................................................... 10-30
PIV Uncertainty ............................................................................. 10-36
CHAPTER 11 Programming Macros .................................................. 11-1
Working with Macros ...................................................................... 11-2
Working with Image Lists ................................................................ 11-3
Working with Batch Files ................................................................ 11-4
CHAPTER 12 Calibrating and Processing PLIF Images .................. 12-1
Capturing PLIF Calibration Images ................................................ 12-1
Processing PLIF Images................................................................. 12-2
Two Camera PLIF Processing ........................................................ 12-8
PLIF + PIV Processing ................................................................. 12-10
Exporting PLIF Data ..................................................................... 12-12
CHAPTER 13 Calibrating, Processing, and Viewing GSV Images . 13-1
Global Sizing Velocimetry Calibration Process .............................. 13-1
Processing Global Sizing Velocimetry Images ............................... 13-2
Viewing Global Sizing Velocimetry Processing Results ................. 13-8
CHAPTER 14 Processing Spray Images ........................................... 14-1
Hardware Setup for Sprays ............................................................ 14-1
Software Setup for Steady Spray ................................................... 14-2
Software Setup for Pulsed Spray ................................................... 14-4
vi Insight 4G Global Image Acquisition, Analysis and Display Software

CHAPTER 15 Micro Particle Image Velocimetry (MicroPIV)............ 15-1
MicroPIV Overview ......................................................................... 15-2
MicroPIV Operating Principles ........................................................ 15-4
MicroPIV Experiment Kit and Example Experiment...................... 15-12
CHAPTER 16 Presenting Data with Tecplot® Software ................... 16-1
Installing Tecplot® Software ............................................................ 16-1
Accessing and Using TecPIV and Tecplot® Software..................... 16-2
Selecting and Reading Files into Tecplot® Software ...................... 16-2
Selecting Velocity Vectors and Modified Vectors ......................... 16-10
Time-Resolved Toolbox ................................................................ 16-25
Histogram Plot............................................................................... 16-28
Scatter Plot.................................................................................... 16-32
Troubleshooting ............................................................................ 16-34
For Further Information ................................................................. 16-35
CHAPTER 17 Enabling Distributed Processing (PIV and
GSV Only) ...................................................................................... 17-1
CHAPTER 18 Developing and Using Plugin Dynamic Link
Libraries (PIV Only) ....................................................................... 18-1
Plugin Definition and Structure ....................................................... 18-1
PluginData and Dynamic Link Libraries (DLL) Plugin Parameters . 18-2
Plugin Development Kit................................................................... 18-3
Installing TSI Plugin Wizards into Visual C++................................. 18-3
Creating a Plugin Using a Wizard ................................................... 18-5
Plugin SDK Files and Classes ...................................................... 18-16
Data Files ...................................................................................... 18-27
APPENDIX A Insight Data Files ........................................................... A-1
Insight Software Filenaming ..............................................................A-1
Insight Software Files ........................................................................A-2
Vector (*.VEC) Files ..........................................................................A-3
Particle (*.PAR) Files ......................................................................A-11
Scalar Data (*.SCA) Files ...............................................................A-11
Statistics List (*STL) File .................................................................A-13
Image Capture Timing (*.tstmp) File ...............................................A-15
3-D Vector Files ..............................................................................A-16
APPENDIX B Using MATLAB® Program ............................................. B-1
Getting Started ..................................................................................B-1
Loading Insight Software Files ..........................................................B-2
Using the Spatial Toolbox .................................................................B-3
Using the Time Series Analysis Toolbox ........................................B-11
Using the Proper Orthogonal Decomposition (POD) Toolbox ........B-30
APPENDIX C Enabling and Setting the Traverse .............................. C-1
Enabling and Setting up the Traverse ............................................. C-2
Index
Reader’s Comments
Contents vii
(This page intentionally left blank)
viii Insight 4G Global Image Acquisition, Analysis and Display Software

About This Manual
Purpose
This manual describes how to use TSI’s Insight 4G™ Global Image
Acquisition, Analysis and Display Software.
Related Product Literature

Please refer to the other PIV manuals that were shipped to you for
additional information.
Submitting Comments
TSI values your comments and suggestions on this manual. Please use
the comment sheet, on the last page of this manual, to send us your
opinion on the manual’s usability, to suggest specific improvements, or to
report any technical errors.
If the comment sheet has already been used, mail or fax your comments
on another sheet of paper to:
TSI Incorporated
500 Cardigan Road
Shoreview, MN 55126
Fax: (651) 490-3824
E-mail Address: [email protected]
ix
x Insight 4G Global Image Acquisition, Analysis and Display Software

CHAPTER 1
Introduction
Insight 4G™ software is a Microsoft® Windows®-based software that is

used for global image capture, analysis and display purposes. It is used for
PIV measurements (global velocity diagnostics), PLIF measurements
(scalar property field) and for global spray diagnostics.
Insight 4G software is a 64-bit Windows® 7 application.
Insight 4G software acquires, analyzes global images (velocity field, or

particle images as well as scalar image fields), and displays the global
properties and associated statistics. The images are generally obtained by
a digital camera.
Insight 4G software automates the process of image capture, analysis and

data validation for each image. This helps you obtain instant feedback,
which makes the setup process for an experiment, fast and informative.
Insight 4G software‘s batch-mode capturing and processing feature allows
you to capture and process large number of images in actual experiments.
Insight 4G software is functionally divided into two parts—an acquisition

and processing part and a presentation part. The first acquires an image
and processes it to obtain various properties such as velocity vectors,
temperature or concentration information or spray geometry. The second
part displays these results which can be enhanced for optimal viewing.
Additional tools such as Tecplot® and MATLAB® programs can also be
used, providing for additional viewing options.
Insight 4G software is a true Global Image capture, analysis, and display

package. Features include:
 Uncertainty analysis for PIV processing to provide confidence level to
PIV measurements
 Dynamic masking on objects to include or exclude vector processing
 Dynamic vortex tracking for the entire flow field
 Per, mid and post triggering to capture desirable images at the correct
occurrence
 Measure microflows (using a microPIV system) for microfluidic and in
microchannels
 Make time-resolved measurements of flow fields (using high-frame rate
cameras and high-pulse rate lasers)
®
Windows is a registered trademark of Microsoft Corporation in the United States and other countries.
Insight, Insight 3G, and Insight 4G are trademarks of TSI Incorporated.
Tecplot is a registered trademark of Tecplot, Inc., Bellevue, WA, USA.
MATLAB is a registered trademark of The MathWorks, Inc.
1-1
 Make global measurement of scalar properties (using PLIF and special
cameras), including species, soot and temperature in combustion
 Make simultaneous velocity and scalar property (temperature,
concentration, species) measurements (using same type of cameras or
using different types of cameras to capture particle image and global
fluorescence fields)
 Make global spray diagnostics measurements to provide sizing and
velocity information simultaneously.
The Package allows the setup, control and operation of the entire
diagnostics system and provides on-line or batch-mode analysis of the
image fields and detailed display of the results.
1-2 Insight 4G Global Image Acquisition, Analysis and Display Software

CHAPTER 2
Getting Started
Be sure to do the following before starting:

 Unpack and check your Insight 4G™ (hereafter referred to as Insight)
software package.
 Check to see if you have the required hardware and software to run the
program.
 Run the installation program.
After completing the above steps, you can start the program.
Note: This manual assumes that you know the basics of using computer
and Windows-based applications such as how to start applications,
how to use your mouse, move and close windows, and other such
tasks. If you need help with these, consult your computer and
Windows® operating system documentation.
Unpacking and Checking Package Contents

The Insight™ software package includes:
 Installation DVD
 Software key—the key connects to the USB port of your computer and
is required to run the program
 User’s Guide (P/N 6004904)
Note: The proper installation of Insight software is determined by the

software key used. Make sure that the software key has a label
indicating the type of Insight software that you will install. If there is
any question on the type of Insight software or if the label on the
software indicates a different type from what you purchased, please
contact TSI immediately.
Carefully unpack and make sure you received all the listed items. If
anything is missing or damaged, contact your nearest TSI sales office or
representative or contact:
TSI Incorporated
500 Cardigan Road
Shoreview, MN 55126
Phone: 1-800-874-2811 (USA) or (651) 490-2811
Fax: (651) 490-3824
E-mail: [email protected]
2-1
Hardware and Software Requirements
Insight software requires or recommends the following hardware and
software:
 Intel Core i3 computer (recommended) with Windows 7 Professional
operating system.
 For Insight 4G software, 64-bit Windows® 7 operating system is
required with Professional the preferred version.
 DVD drive
 Minimum of 6 GB of RAM
 Disk drive with at least 500 GB
 Video card set for 32K colors or more
 17-in. monitor; Insight software works best with a 19 in. or larger
monitor.
Installing Insight Software

To install Insight software:
1. Insert the DVD in the appropriate drive.
The installation wizard starts automatically. If installation does not start
automatically, double-click setup.exe.
2. Click Install button and follow the instructions that are displayed on
your screen to complete the installation process. The installation of the
software may take up to 30 minutes due to the various types of drivers
and control software. Be sure to wait until the software is completely
installed. If you discontinue in the middle of the installation process, the
entire software will have to be reinstalled again.
3. Refer to the Installation Guide for other steps of the software install.
The latest software Patch is available on the TSI FTP site. It is always
recommended that you check the FTP site and install the latest Patch
version. There is no extra charge for installing the latest patch version
of the software. Follow the instructions to install the latest Patch
version.
Starting Insight Software

To start Insight software:
Double-click the Insight icon on your desktop or from the Windows® task
bar, choose Start | Programs | TSI Insight 4G | Insight 4G. In order to
run Insight software without the splash screen, pass it the command-line
option NO_SPLASH. The NO_SPLASH option will speed up launch time,
especially when running the computer using Remote Desktop.

CHAPTER 3
Getting Acquainted
with Insight Software
Insight™ software provides many features to help you make flow and spray
measurements. This chapter acquaints you with the Insight interface to
give you an overview of the software and its capabilities. Details on each
option and task performed in Insight software are provided in subsequent
chapters.
The Insight Desktop

When you start Insight software, a desktop with menu, tool bar and tab
options is displayed. As you use Insight software to capture, process, and
display images, you will use several of these options. Review the following
information to become acquainted with its many features and capabilities.
Figure 3-1
Insight Software Desktop
3-1
Insight Menu Options
The bar at the top of the Insight application window contains the following
sets of drop-down menus.
Experiment/Run
The options in the Experiment menu allow you to open a new experiment
or a run or edit a previously-saved experiment or run. It also allows you to
sort and refresh the experiment tree display in the control panel. Following
are brief descriptions of these commands.
Refer to Chapter 6, “Viewing, Enhancing, and Displaying Image Files,” for

details.
Option Description
New Experiment Creates a new experiment.
Edit Experiment Opens an existing experiment for editing.
New Run Creates a new run.
Edit Run Opens an existing run for editing
Sort Tree Sorts and displays your experiment and data files
according to a set of specifications that you choose. See
“Specifying the Sorting Order for Data Files” for details.
Refresh Tree Refreshes and presents an updated tree structure.
Exit Exits the Insight program.
Tools
The Tool menu provides tools to specify calibration parameters, capture
PIV and PLIF calibration images, specify hardware components, and
program macros that can automate your experiments and runs. Following
are brief descriptions of these commands.
Option Description
Insight Setup Sets up experiment directory and computer for distributed
processing.
Hardware Setup Sets up parameters for the components to be used with
the Insight software. See “Specifying Hardware
Components in Your Experiment” in Chapter 4.
Component Setup Sets up individual hardware components (e.g., individual
cameras, lasers, etc.)
Image Tools Provides tools to modify images not captured by Insight
software so they can be used by Insight software.
MATLAB Spatial and Time Series Analysis.
Proper Orthogonal Decomposition.
View and analyze results with MATLAB applications.
Capture Captures images for Perspective Calibration. See “Step C:
Perspective Cal. Recording Calibration Images” for details.
Images

Window
The Window menu option allows you to control and arrange the open
windows in the display panel. The following options are available:
Tool Bars Description
Cascade Cascades all open windows.
Tile Horizontal Tiles all open windows in a horizontal format.
Tile Vertical Tiles all open windows in vertical format.
Arrange Icons Arranges the icons so they can be properly displayed.
Help
The Help menu option allows you to access the instruction and operational
manuals available for Insight software. If you click on the “Insight 4G” icon,
it also tells you the various functions which the software should be
activated. This is a good way to check whether the dongle key functions
properly.
Insight Tool Bar

Insight software makes many options easily accessible to you through the
toolbar. This tool bar appears under the main menu bar and contains a row
of buttons and tools to allow you to quickly choose commonly used
commands and options.
Option Description
New Run
Creates a new run. Performs the same function as the
corresponding command in the Experiment/Run menu option. See
“Creating a Run” for details.
Edit Run
Opens an experiment. Performs the same function as the
corresponding command in the Experiment/Run menu option. See
“Editing an Experiment or Run” for details.
Sort Tree
Brings up the dialog box to allow you to specify how to sort and
display files in the Experiment tab panel. Performs the same
function as the corresponding command in the Experiment/Run
menu option. See “Specifying the Sorting Order for Data Files” for
details.
Tecplot
Starts the Tecplot® utility which allows you to view Insight vector
files and compute flow properties. See Chapter 16, "Presenting
Data with Tecplot Software" for details.
Turns the laser off. Performs the same function as the laser off
button available on the tab on the control panel. See Chapter 7,
"Capturing Images: An Overview" for further information.
Getting Acquainted with Insight Software 3-3

Option Description
Runs the macro listed in the drop-down list box next to the button.
This list contains all the saved macros in the current experiment
that were programmed using the Visual Macro Programming
command available through the Tools menu option. See Chapter
11, "Programming Macros" for details.
Stops the macro that is running. See Chapter 11, "Programming
Macros"” for details.
Control Panel
Insight control panel offers many options grouped into the following tabs:
Tab Description
Displays experiment and run data files specified and sorted
through the Sort Tree dialog box. See Chapter 5, "Creating
and Managing Experiment and Data Files".
Allows you to specify capture settings for various
components in your experiment, save these settings and
start and stop image capture. See “Chapter 7, "Capturing
Images" for detailed information.
Allows you to specify Application, Preprocessing,
Processing, Post-processing, Calibration, and Mask
Processing.
Allows you to specify Spray analyses and processing
settings. See Chapter 14, "Processing Spray Images" for
details.
Display Panel
Insight software displays processed image files in the display panel. To
view these files, double-click on files in the Experiment tree or drag and
drop them on the display panel. Each file opens in a separate window with
controls offering more functions. Refer to Chapter 6, “Viewing, Enhancing
and Displaying Image Files” for details.
Status Bar
The status bar on the bottom of the Insight desktop provides status on the
hardware used in the experiment. To get an updated status of a new
hardware component in a system, click on the component in the status bar.
The status bar also lists the name of the current experiment and run
indicating where the data is being saved or retrieved from.

CHAPTER 4
Setting up Insight
Software
Before acquiring and processing PIV, PLIF, and spray images, perform the
following tasks:
 Specify the directory to which the acquired data in an experiment is
stored.
 Specify the hardware components in your experiment.
 (Optional) Set up Insight™ software to perform distributed processing
over a network of computers so that processing time for PIV images
can be reduced.
 (Optional) Set up and configure dual monitors to help you visualize
your images on more than one desktop.
The following sections provide details on how to perform these tasks.
Specifying the Experiment Folder

Insight software allows you to save your experiment data to any directory
and computer on your network.
Note: To avoid causing errors in Insight software, it is recommended that

the experiment folder be used exclusively by Insight software. Do not
use it to store files from other applications.
To Specify an Experiment Folder

1. From the main menu of Insight software, select Tools | Insight
Setup... The Setup Insight dialog box appears.
2. In the Directory for Experiments box, type in the directory name where
you want to save the data files or click to browse through your
computer or network to select the designated folder. See "Control
Panel" for more information.
4-1
Specifying Hardware Components in Your
Experiment
Insight software works with many types of cameras, lasers, and
synchronizers. Before acquiring images, you need to specify which models
or types of hardware components are being used in your experiment and
how they are connected.
To Specify the Hardware Components in your

Experiment
1. From the main Insight menu, select Tools | Hardware Setup... The
Hardware Setup dialog box opens.
2. Perform the following:
a. Select the appropriate frame grabber from the following list of
supported models for the first camera in your experiment:
Supported Frame Insight 4G
Grabber Models Software
Phantom Gigabit Ethernet X
Photron Gigabit Ethernet X
Xcelera-CL PX4 X
Xcelera-CL PX8 X
Andor USB X
b. The list of cameras will automatically reduce to those supported by

your selected frame grabber. Choose a camera appropriately.
c. Once the camera is selected, the list of ports will automatically
reduce to those supported by your frame grabber/camera pair.
Choose a port appropriately (refer to the appropriate device
manual for an explanation of these ports).
d. Select the type of application that the camera will be used for.
Possible applications include: PIV Mono, PIV Left, PIV Right, PLIF,
and GSV.
e. Repeat a through d for the other two frame grabbers, if they are
being used. There are cameras which use the same frame grabber
for stereo PIV arrangement. If this is the case, select the camera
type based on the SAME frame grabber. Designate the appropriate
PORT and APPLICATION for the camera.
f. If another or different frame grabber is used for the second
camera, select the “Frame grabber Model” for the camera using
the second Frame grabber selection box.
g. Specify the synchronizer being used and specify the port it is
connected to, in the adjoining box.
h. Select the laser being used.
i. Check Traverse Installed, if a traverse is installed.

j. Check IO Board Installed, if an I/O board is installed in the
computer and if the I/O board needs to be activated for analog
data input.
k. Select the Auxiliary Device installed from the drop down list.
3. Click OK.
Next, you would also need to setup the parameters for these selected
components, as described in "Setting Hardware Component Parameters".
Using Dual Monitors

Insight software is designed to take advantage of dual monitor systems. If
your computer has this capability, this feature allows you to connect and
use more than one monitor on to your computer to provide a wider and
more flexible display surface for your experiment results. It also allows you
to dock the control panel on either the right or left side of your monitor to
make it easy to drag and drop the files from the experiment tree to the
display panel on the other monitor.
To Use Two Monitors

1. In the Setup Insight dialog box, check Floating Image Window.
Additional options become available.
2. Depending on your preference and the placement of the monitors,
check Dock Right or Dock Left to dock the control panel on either the
left or the right end of the monitor.
3. Follow the instructions provided in your Windows® manual on how to
set the dual-monitor option.
4. Start Insight software. You can move any child windows in the display
panel of Insight software to be displayed on the second monitor.
Setting Up Insight Software 4-3


CHAPTER 5
Creating and Managing
Experiment and Data
Files
Insight™ software’s Experiment/Run menu options and the Experiment

control tab allow you to create your experiment and run folders to store
your experiment settings and data files as well as organize and access
your raw and analyzed data files.
This chapter describes how to create and work with experiments and how
to use the features provided in the Experiment tab to sort, organize and
access the settings and data files for the experiments and runs.
Creating Experiment and Run Folders

An experiment is a collection of shared settings and run data. The shared
settings consist of information such as capture setups, processing setups,
image conditioning setups, ROI (Region of Interest) files, and macro files.
The settings are shared across runs with in the experiment. After you
create an experiment folder you need to create runs within each
experiment before you can capture any images for the experiment. This
section discusses how to create an experiment folder and runs within it.
Creating an Experiment
To create an experiment:
1. Choose New Experiment from the Experiment/Run menu. The New
Experiment Information dialog box appears.
2. Specify a name for your experiment in the Experiment Name box and
provide any additional information in the Notes section. The Notes
section allows you to provide more details of your experiment. Insight
software creates a folder by that name in the directory that you created
when setting up Insight software. (See Chapter 4, “Setting up Insight
Software.”) It also creates an experiment folder with two subfolders in
the Exp. Tree tab in the control panel. The experiment folder is marked
with the icon and by default is designated as current. Two
subfolders—a file with information specified for the experiment and a
folder called Settings are created. See "Accessing the Experiment
Tree" in this section for details.
5-1
Creating a Run
To create a run within an experiment:
1. Make sure the experiment you want to create the run for is current.
(See "Setting an Experiment or Run as Current" in this section.)
2. Choose New Run from the Experiment/Run menu or from the tool bar.
3. The New Run Information dialog box appears.
4. Specify a name for the run. Provide any additional information in the
Notes section. The Notes section allows you to input more information
of this particular run. This is the best location to provide the details of
the Run so that you can keep track of the different conditions for each
Run. Insight software creates a folder by the name you specified in the
directory that you created when setting up Insight software. It also
creates a run folder under the selected experiment in the Exp. Tree tab
in the control panel. The run folder is marked with the icon and by
default is designated as current. See "Accessing the Experiment Tree"
in this section for details.
Editing an Experiment or Run

To edit the notes for a previously created experiment or run:
1. Choose Edit Experiment or Edit Run from the Experiment/Run menu.
The Experiment Information or Run Information dialog box appears.

2. Edit the information in the Notes section and click OK.
Managing Insight Files

As mentioned earlier, when you create a new experiment or run, Insight
software creates the experiment folders files in the directory that you
created when setting up Insight software as well as in the Experiment Tree
tab within the Insight software interface. The Experiment Tree tab allows
you to manage, access, and organize all your data files easily. The
following sections provide details on the various tasks that can be
performed using the Experiment Tree command options.
Accessing the Experiment Tree

Click on the control panel. The experiment tree display
opens. The following shows an example with a single expanded
experiment folder and its contents.

Figure 5-1
Accessing Experiment Tree
The following sections describe the various tasks that can be performed
from the Experiment Tree.
Specifying the Sorting Order for Data Files

Specify the order in which data files should be displayed using the Sort
Tree option.
To Specify the Display Order of Data Files

1. Choose Sort Tree from the Experiment/Run menu or from the tool bar.
The Experiment Tree: Sort Order dialog box appears.
Note: To edit the sort order, right-click on the experiment name in the
tree and select Edit Sort Order.
2. Select the order in which you would like files to be sorted by clicking
the preferred display option in each of the boxes. The choices include:
Option Description
Capture Select to sort files in the order of in which they were
captured.
Traverse Select to sort files in the order of their specified
traverse setting.
Delay Select to sort files in the order of their specified delay
time.
Process Select to sort files in the order of their specified
process.
File Type Select to sort files in the order of their file types.
Creating and Managing Experiment and Data Files 5-3

3. Choose the file fields to be displayed in the file name from the
following:
Option Description
Show Traverse Select to display the traverse field setting in the
filename.
Show Process Select to display the process field setting in the
filename.
Show Delay Select to display the delay time field setting in the
filename.
Show Process Select to display the process field setting in the
filename.
Show Hardware Select to display the hardware field setting in the
filename.
4. Click OK. The data files, when available, are sorted and displayed in
the specified order in the Raw Data and Analysis folders. The following
figure illustrates the fields in the filename when they are all selected.
Figure 5-2
File Naming Diagram
Setting an Experiment or Run as Current

When you select an experiment or run and set it to be current, all the data
files are saved in the current run and settings are saved and accessed
from the current experiment folder.
To Specify an Experiment or Run as Current

1. Right-click the experiment or run folder you want to make current.
2. Choose Set Experiment as Current or Set Run as Current. The icon
color changes to a bright yellow indicating that it is the current
experiment or run. The status bar as well as the title bar displays the
name of the experiment set as current.
Renaming an Experiment/Run Folder

To rename an experiment or run:
1. Right-click the experiment/run folder you want to rename.
2. Choose Rename Experiment or Rename Run. The Rename
Experiment box or Rename Run box appears.

3. Enter a new name and click OK.
Deleting an Experiment/Run Folder

To delete an experiment or run folder:
1. Right-click the experiment or run folder you want to delete.
2. Choose Delete Experiment or Delete Run. A warning message
appears alerting you that changes are permanent.
3. Click OK.
Expanding or Collapsing the Experiment Tree

Expand an experiment tree to display the contents of the entire experiment
and run folders and subfolders or collapse it to hide the subfolders.
To Expand an Experiment Folder

 Single-click + next to the experiment or run folder or right-click the
experiment or run folder and select Expand Tree. The tree structure
expands and displays the contents of all the folders.
To Collapse an Experiment Tree

 Single-click - next to the experiment or run folder or right-click the
experiment or run folder and select Collapse Tree. The tree structure
collapse and displays only the top-level folders.
Creating and Managing Experiment and Data Files 5-5


CHAPTER 6
Viewing, Enhancing,
and Displaying Image
Files
The display panel in Insight™ software allows you to drag and drop images
from the Experiment tree and then using the many features available in the
display panel, enhance the images so that you can process them, view
them or select a certain region of interest. Insight software also provides
you with image information, controls to scroll through or animate the data,
and the means to export images in different formats.
This chapter describes the display panel features and how to do the
following:
 Zoom in or out of an image for optimal viewing.
 Turn vector display on or off for speed (Turbo mode).
 Create an image list.
 Select entire images or use cross-hair marks to help with processing
and calibration.
 Modify output lookup tables to enhance images for optimal viewing.
 Edit and change the appearance of vectors, remove, apply, and view
vectors.
 Get statistical information on images and vectors, display image
information bar and make changes to the display.
 Export images.
 Animate images.
6-1
Display Panel Features
The following figure illustrates Insight software‘s features.
Figure 6-1
Display Showing Features
The following describes these display features.

Zooming In and Out of Image Files
When you open a data file, Insight software brings up the image at full
resolution—each pixel is mapped to a pixel on the screen and often the
image is too large to be seen in its entirety. To view the entire image you
may have to zoom in or out.
To Resize, Zoom In or Out of Image Files

1. Open and drag an image file on to the display panel. The display
window opens and the display features become available.
2. From the Tool Bar, click Size. The following options are available.
Option Description
Fit in Window Sizes the image to fit in the display window.
Full Size 1:1 Displays image at actual or full resolution.
Zoom In 2:1, 3:1 4:1 Zooms in to provide a closer look at the image file.
Zoom Out 1:2, 1:3, 1:4 Zooms out to provide a larger or entire view of the
image file.
Note: Zoom into the image by right-clicking and dragging the mouse.
Display While Processing – “Turbo Mode”

This button controls if the vectors are displayed while processing. In
Display While Processing mode each vector is added to the display as it is
computed.
The display of each vector takes time. Turning the Display While
Processing off can increase the processing speed by up to 2 times.
When processing a long sequence of images, the recommendation is to

turn Display While Processing on for the first few vector fields, so you
can see the processing is as you intended. Once you have confirmed the
processing, the Display While Processing can be turned off to increase
the processing speed.
Viewing, Enhancing, and Displaying Image Files 6-3

Create Image List
An image list can be used to open a group of image form the settings
folder, and it is used to define the images for batch processing. The image
list can be created in the Experiment Tree by selecting the images then
right clicking and selecting Create Image List.
Figure 6-2
Experiment Tree
Image List
Creation
An image list can also be created by opening a group of images then

clicking the Create Image List button.
With either creation process the image list name dialog will open. The
saved image list is stored in the experiment settings folder.
Figure 6-3
Image List Name Dialog

Applying Crosshairs
Applying crosshairs shows the center of the image and is useful for
aligning cameras when you are performing calibration in a PIV experiment.
See Chapter 8, Step A “Overlapping the Camera Fields of View with the
Calibration Target” for details.
To Turn on Crosshairs
From the main menu bar of the display panel select Tools | Cross hairs. A
check mark appears on the Crosshairs option and crosshair marks appear
on the images.
To Turn off Crosshairs

From the main menu bar of the display panel select Tools | Cross hairs.
The check mark from the Crosshairs option and the crosshair marks on the
images are removed.
Applying Grids
Applying grid shows a grid on top of displayed images. The grid can be
configured automatically or manually with the options listed.
To Turn on Grid Option

From the grid option, check Auto Apply. A check mark appears on the
Auto Apply and a grid appears on the images.
To Turn off Grid Option

From the grid option, un-check Auto Apply. Grid on the images is
removed.
Selecting Entire Images

The Select Entire Image option is useful when you are processing captured
images and want to make sure the entire image has been selected.
To Select the Entire Image

From the main menu bar of the display panel select Tools | Select Entire
Image. A check mark appears on the Select Entire Image option and the
entire image is selected.

Enhancing Image Displays by Modifying Look Up
Tables (LUT)
Insight software allows you to manipulate and display the Look Up Tables
(LUT) of the graphics board in your computer to enhance the image
display. Although the LUT manipulates the pixel intensity values displayed
on your computer monitor, it does not affect the values in the stored
image file.
Two look-up table options are available. One for PIV images and the other
for PLIF images, as the latter tend to be darker.
Insight software captures 8-bit, 12-bit and 16-bit images, depending on the
camera that is used for the image capture. For 8-bit images, each pixel in
these images has a grayscale assigned to it, ranging from 0 to 255, with
0=black and 255=white. You can also use pseudo colors and assign a
color ranging from 0-255 with 0=black and 255=red. When an image is
initially displayed on your computer screen, the graphics board in your
computer displays this image based on that number. Use the LUT option to
modify each pixel intensity—for example, alter image brightness or the
contrast or both to bring up the details.
For 12-bit images, there are a total of 4096 intensity levels; hence, it is
better to show the images in pseudocolor so that the contrast of the images
can be seen more easily.
To Use the LUT Option

1. From the display panel menu bar, select Tools | PIV LUT or Tools |
PLIF LUT. The PIV LUT or PLIF LUT dialog opens.
Figure 6-4
PLIF LUT Table

2. Use the following LUT options and controls to manipulate the image
displays:
Option Description
Displays the current settings. Use this to preview
the effect of changes made through controls and
options available in this dialog box.
Slide these tabs for each color or grayscale value
to make gross changes. Use the bar to preview
the settings.
Use these controls to make minute changes to
the color or gray-scale values.
Pseudocolor Select this check box to display colors instead of
grayscale values.
Auto Max Use for extremely dark images. Insight software
automatically finds the highest intensity level in
the bit map and assigns a top intensity value to it.
Linear Click to apply the linear function on the color
range to spread the color or grayscale range
linearly.
Log Click to apply the logarithmic function on the color
range.
Change Colors Click to change text labels or color for the color
bands or saturation color. The Change Colors
dialog appears.
Remember Check box to save the settings and reload them
Settings next time this image is opened.
Load LUT Click to browse for and load a LUT file containing
color values, text labels for bands, and slider
settings.
Save LUT Click to save the current slider settings, color
values, and text labels to a LUT file.
Saturation color Displays the color selected to indicate saturation
levels.

Editing, Displaying, and Clearing Vectors
Insight software’s vector editing options allow you to setup and change
vector appearance, to apply vectors automatically and to clear vectors on
your PIV images.
To Setup Vectors
1. Click on Tools | Vector Setup. The Vector setup dialog box opens.
2. Select the appropriate option or options. The following table describes
the Vector options that affect the appearance of vector files:
Option Description
Vector Settings Changes the settings of the vector arrows. The color of
Good, Bad, and Interpolated vectors can be set
independently. Whether they are displayed on the
images can also be checked/unchecked.
To change the color:
1. Right-click on box. A color palette is displayed.
2. Select a new color to represent vector arrows.
1. Click OK.
Scale Changes the length and width of the vectors. Use the
arrows to scroll up and down to select the desired scale.
Screen Updating Select the number of vectors to display from the following
selections.
Draw All: Displays all the vectors.
Draw 1/2: Displays half of all the vectors.
Draw 1/4: Displays a fourth of all the vectors
Draw 1/8: Displays an eighth of all the vectors.
3. Click Apply to see the effect of the change and OK to save the
changes. Cancel discards the changes.
Getting Information on Images

Insight software also provides intensity levels in the form of statistics and a
histogram. It also provides additional information on the images. Following
describes how to access and use these features.

Image Statistics
The Image Statistics feature provides a table displaying intensity levels.
These statistics can be printed and exported to other applications, copied
to a clipboard to be inserted into another application such as Microsoft®
Word.
To Display Image Statistics

1. Drag and drop or double-click the desired image file from the
Experiment tree to the display panel.
2. From the main menu of the display panel, select Image

Statistics. The statistics table, as shown in the following figure, is
displayed.
Figure 6-5
Statistics Table

To Export Image Statistics
1. Select Print/Export from the statistical table menu bar. The Table
Print/Export dialog opens.
2. Make your selections from the following export options:
Option Description
CSV Select to export the table in comma separated
value format.
Excel Select to export the table into a Microsoft® Excel®
spreadsheet format.
For MATLAB Select to export the table into MATLAB® program. See
Using MATLAB for details.
3. Specify where you want to send the exported file. Click to browse
to the appropriate drive.
4. Click OK.
To Print Image Statistics

1. Click Print/Export from the statistical table menu bar. The Table
Print/Export dialog opens.
2. (optional) Select the Print Setup box if you would like to specify your
printing preferences before printing.
3. Click Print. If you selected Print Setup, the printer dialog opens.
Specify your printing preferences and click OK.
To Send the Image Statistics Table to a Clipboard

1. Click Copy to Clipboard from the statistical table menu bar.

Image Histogram
Image histogram feature provides a histogram displaying intensity levels.
To Display the Image Histogram


Histogram. The image histogram, as shown in the following figure, is
displayed.
3. Click Y-Axis Label button to change the text of the Y-Axis label.
Figure 6-6
Image Histogram

Image Information
Image information provides a histogram displaying intensity levels.
To Display the Image Information


Information. The image histogram, as shown in Figure 6-7, is
displayed.
Figure 6-7
Image Information Screen

Vector Information
Get additional vector information in the form of detailed statistical table or a
histogram.
To Display the Vector Statistics

Experiment tree on to the display panel.
2. From the main menu of the display panel, select Vectors |

Open Left Vector Table or Open Right Vector Table. The vector
statistical table, as shown in the following figure, is displayed
(Figure 6-8).
Figure 6-8
Vector Statistical Table

To Display the Vector Histogram
2. From the main menu of the display panel, select Vectors |

Open Left Vector Histogram or Open Right Vector Histogram. The
vector histogram, as shown in Figure 6-9, is displayed.
Figure 6-9
Vector Histogram
Line Profile
To plot line profile of images, open the Line Profile dialog box. Both vertical
and horizontal line profiles can be displayed. Y-axis label can be
customized to fit your needs.

Exporting Images
The Export option available from the Tools | Export Images… dialog.
Export Images allows you to export images from Insight software into other
applications. These images can be exported in three formats: .JPG, TIF,
and BMP. Exported files can be viewed using any imaging software
application.
Figure 6-10
Export Images Dialog
To Export Images
1. Drag and drop or double-click the desired image files from the
2. From the main menu of the display panel, click . The Export
Images dialog box is displayed.
3. In the Images Loaded box, select the images you want to export. Use
the Select All or Select None buttons to make quick selections.
4. Select the format in which you want to export. Choices include: BMP,
JPG, or TIF.

5. Select the Data Mode option:
Option Description
Preserve Data Export the raw data as it is captured.
Apply LUT Check to use values from the current LUT and
export it exactly as it is displayed.
Include PIV vectors Select to have the vectors displayed on the
images to be exported on top of the images.
6. Specify where you want to send the exported file. Click to browse
to the appropriate drive.
7. Click Export.
Animating Images
The animation feature available on the display panel menu bar, allows you
to view all the loaded files in the form of an animation.
To Animate Images
1. Drag and drop or double-click the desired image files from the
2. From the main menu of the display panel, click . Each loaded
image file is displayed in the form of an animation.

CHAPTER 7
Capturing Images
Insight™ software offers options to capture images for a variety of optical

measurement systems, including: PIV, StereoPIV, MicroPIV, PLIF,
Simultaneous PIV and PLIF, GSV, Spray Analysis, and Flow Visualization.
The setup and controls for capturing these images are grouped together on
a tab in the control panel.
This chapter describes how to specify the capture settings for the
components for your experiments, how to save them for future use, and
how to adjust and manipulate other settings for optimal data acquisition.
Capturing Images: An Overview

The Capture tab on the control panel provides options to:
 Set parameters to capture images
 Perform image captures, and
 Save image capture settings for future use.
Note: All these tasks can be performed dynamically anytime during an
experiment.
The following gives a quick look at the steps involved in capturing images.
Details on each of these steps are provided later in this chapter.
Set Parameters to Capture Images

1. Click tab on the control panel. The capture options
appear.
2. Select the application for which the image is being captured. Click to
select PIV, PLIF, PIV-PLIF, or GSV.
3. Set up the hardware components parameters. Click Component
Setup… Refer to “Setting Hardware Component Parameters” for
details.
4. Set up the traverse, if you are using one. Click Traverse Setup Refer
to “Enabling and Setting Traverse” for details.
5. Set up the timing parameters for your capture. Click Timing Setup…
Refer to “Setting Timing Parameters” for details.
6. Select the exposure mode in the Exposure selection box. Refer to
“Selecting the Exposure Mode” for details.
7-1
7. Select the capture mode in the Capture selection box. Refer to
“Selecting the Capture Mode” for details.
8. Specify name for the capture files.
9. Click and select the laser power for each of lasers. See “Selecting
Laser Power and Pulse Energy.”
10. Click Laser On to turn on the lasers.
Capture Images
Click Capture.
Capture Images Post-Trigger

If Software Trigger is enabled (see "Selecting the Capture Mode" for more
information), click to start post-trigger capturing.
Save Capture Settings and Images to Disk

1. Save capture settings, load previously saved capture settings by using
the Saved Capture Setup options. See “Saving, Loading and Deleting
Image Capture Settings” for details.
2. If you had not already selected the option of saving RAM images to the
hard drive when selecting the capture mode (see “Selecting the
Capture Mode”), click Save RAM Images. Images are saved in the
Experiment folder.
3. Click Laser Off to turn off the lasers.
Refer to the following sections for detailed information on each of these

steps.
Setting Hardware Component Parameters

Before capturing images, you need to specify and set values for the
hardware components that you had specified earlier. The following
describes how to set the parameters for the following hardware
components:
 Synchronizer
 Camera
 Laser
The parameters are set by selecting Tools | Component Setup from the
Insight main menu. This causes the Component Setup dialog to be
displayed. The summary tab in this dialog box displays the selected
models for each of these components.

Setting Synchronizer Parameters
This procedure involves setting up the parameters for the synchronizer that
was selected during the Insight software setup. See “Specifying Hardware
Components in Your Experiment.”
To set the synchronizer parameters:

1. From the Tools menu, select Component Setup.
2. Select the Synchronizer Setup tab in the Component Setup dialog
box. The Synchronizer Setup dialog is displayed.
The synchronizer uses TTL signals to communicate with most of the

other system components. These TTL signals come in two polarities +
and –. Each component has a standard polarity setting that is set when
the component is selected. You may need to set the polarity if you are
supplying an external trigger to the system
3. Select the polarity for each of the components listed under Polarity.
Make the selection based on the following descriptions.
Option Description
+ Polarity Normally at ground and goes high to trigger another
device. Timing is set for the rising edge.
- Polarity Normally high and goes to ground to trigger another
device. Timing is set for the falling edge of the signal.
Version Indicates the version number and date of the
firmware used in the synchronizer. The version
number is helpful when checking with the factory to
see if the firmware is up-to-date, especially when
problems occur on the synchronizer.
Send All Sends all the synchronizer settings for the software
to the synchronizer. Under normal circumstances, it
is not necessary to use this button. Any changes in
the settings are sent to the synchronizer
automatically. Use this button mainly to confirm if the
parameters were transmitted.
4. Select Laser Setup tab to setup the laser parameters, described next.
Setting Laser Parameters

This procedure involves setting the parameters for the laser that was
selected during the Insight software setup. See “Specifying Hardware
Components in Your Experiment.”
To set the laser parameters:

1. On the Insight main menu, select Tools | Component Setup. Select
the Laser Setup tab in the dialog box.
2. Make appropriate selections for the selected laser in the Laser Setup
tab. “Laser Setup Parameters” table provides detailed descriptions.
Capturing Images 7-3

Laser Setup Parameters
The following describes the parameters for all types of lasers. These
parameters need to be specified in the respective laser setup dialog boxes.
Option Description
Model Displays the model of the selected laser. Displayed here
for reference only. When the laser model is specified in
the hardware setup menu, the default values for all of the
other parameters are also selected.
Laser Frequency (Hz) Displays the flashlamp frequency and allows you to
change it. A laser gives the most energy per pulse and
best beam quality when it is run at the designed
frequency.
Frequency: Refers to the flashlamp frequency, which
is the rate at which the flashlamps are pulsed when the
External Trigger box is checked.
In most cases, the flashlamp frequency should be set at
the Maximum Flashlamp frequency value where the laser
was optimized.
Minimum Frequency: The minimum flashlamp
frequency for the laser. Some lasers will not fire if pulsed
below this frequency.
Maximum Frequency: The maximum flashlamp
frequency for the laser. The laser will not fire with a
flashlamp frequency above the maximum flashlamp
frequency.
3. Select Camera 1 Setup tab to setup the camera parameters.
Setting Camera Parameters

This procedure involves setting up the parameters for the cameras that
were selected during the Insight software setup. See "Specifying Hardware
Components in Your Experiment".
To set the camera parameters:

1. From the Tools menu, select Component Setup. Select the Camera 1
Setup tab in the dialog box. Depending on the type of camera that was
specified, an appropriate camera setup dialog box appears.
2. Make appropriate selections for camera one. See “Camera Setup
Parameters” table for detailed descriptions.

Camera Setup Parameters
The following lists and describes the parameters for all types of cameras.
These parameters need to be specified in their respective camera setup
dialog boxes.
Option Description
Regular camera models setup
Model and Version Displays model and version of the selected camera.
Resolution Displays the number of pixels in the selected camera and
the spacing between pixels.
Pixels X Y: The number of light sensitive pixels on the
camera.
Spacing (µm): Displays the distance between pixels on
the CCD chip. This value is used in computing the Pixel
Aspect Ratio when converting pixel values to velocity
measurement values. Refer to Chapter 8, “2-D Calibration
Process” for information on the Pixel Aspect Ratio.
Exposure Frame Rate (Hz): Displays the camera’s frame rate,
which is fixed.

Option Description
Digital Gain Enter a value to increase the image brightness level. The
value of the digital gain will be multiplied times each pixel
intensity to adjust for varying levels of environmental
illumination. For example, a gain of 2 will double each
pixel intensity.
Caution; Setting the gain too high will saturate pixels and
result in the loss of intensity information. This happens
because the number of bits allocated for each pixel
remains the same regardless of the gain level. The default
value is 4.
Refer to your camera manual for further details.

Test Image Select box to generate a test image that you can use to
verify your camera connections.
Flip Image Select box to horizontally flip the images that are being
captured. Also see “Stereo PIV Camera Location” in
Chapter 9.
Timeout (ms) Specify a time, in milliseconds, for the camera to wait to
get an image. If the camera does not capture an image
within this specified timeout setting, a timeout message is
displayed.
Trigger Directly Some cameras have the option to either be triggered
through the frame grabber cable, or be triggered directly
from an external source. Check this box to change the
default setting.
Capture Dark Image Some cameras have the ability to acquire an image and
store it in memory, and will subtract off this image during
readout. This image is known as a Dark or Reference
Image, and capturing one with the exact timing, gain, and
area of interest / binning settings will allow for the clearest
and cleanest pictures possible.
Caution
Be sure to put a lens cap over the camera before
acquiring this image.

Option Description
Binning Displays the current binning settings of the camera. Some
cameras allows for various selectable binning options.
Select the desired binning value from the drop down
menus if they are available.
Binning an image simply means combining pixels. This is
done one of two ways. Some cameras allow for control of
both types of binning individually, others require both to
be done simultaneously.
Horizontal Binning: Typically horizontal binning is
typically done by summing pixels together to obtain an
average of the two (or four or eight, etc) and reading out
that value. This method does not affect the frame rate, but
does reduce the resolution horizontally by whatever the
value that is set.
Vertical Binning: Typically vertical binning is done by
reading in multiple lines on one clock, causing the pixels
that are read out to be a summation of pixels. This will
increase the brightness of the image (as these values
aren’t averaged, simply summed), and will also
sometimes increase the possible frame rate of the
camera, as less total lines need to be readout for image
acquisition. This may not be the case for all cameras that
have binning. This will reduce the vertical resolution by
whatever value that is set.
Pixel Clock Displays the pixel clock value of the camera.
Tap Count Displays the amount of taps the camera is currently using
to readout an image. A tap is a channel that brings in
pixels serially, so more taps allow for faster image
acquisition.
Balance Taps This button, when enabled, will read the gain settings of
one tap of a multi-tap camera, and set the corresponding
taps to those exact gain values.
Caution
This should only be used in emergencies when protective
camera settings cause an extreme imbalance between
taps (where one square section of an image is clearly
much brighter or darker than the rest).
Fan Control Select cameras with built-in fans allow for control of fan
settings. This lets you control those settings.
Bits Per Pixel Displays the current amount of bits per pixel that the
camera is reading out. Some cameras may allow this
value to be changed. If this is the case, a secondary
option will appear in the drop down list.
3. Repeat step 2 for other two cameras, if they are being used.
4. Click OK to have the changes accepted and to exit the Component
Setup dialog box.

Enabling and Setting Traverse
If you have a traverse in your experiment and you checked Traverse
Installed in the Hardware Setup dialog, you need to enable the traverse
and set parameters of it. Refer to Appendix C, “Enabling and Setting up the
Traverse” for details.
Setting Timing Parameters

Depending on your application and the exposure mode selected, you need
to adjust the timing for the selected hardware components to help capture
the desired images. All the options to set timing are available in the
Capture Timing Setup dialog box. This dialog box also provides a dynamic
graphical display of the capture timing values to help you adjust the
synchronizer and laser pulses with the image capture cameras. See
“Viewing and Adjusting Timing Parameters.”
To Setup Timing for Image Captures

1. From the Capture tab, select Timing Setup. The Capture Timing
Setup dialog box appears (Figure 7-1).
Figure 7-1
Capture Timing Setup Screen

2. Make appropriate selections. “Timing Setup Parameters” table
provides detailed descriptions.
Timing Setup Parameters

The following describes the parameters in the timing setup dialog box.
Option Description
PIV Frame Mode Defines the number of frames the camera captures when
it receives a trigger from the synchronizer. Select one of
the following:
Single: The camera acquires a single-frame image.
Straddle: The camera acquires two consecutive single-
exposure images.
Pulse Rep Rate (Hz) Select the pulse repetition rate for the laser. This value
specifies the timing from the start of one laser pulse
sequence to the start of the next laser pulse sequence.
Laser Pulse Delay Enter a value for the pulse delay time.
Pulse delay time is the amount of time to wait from the
start of a pulse sequence to until the first laser pulse.
The pulse delay is the parameter that is adjusted to pulse
the laser precisely towards the end of the first frame.
For Nd:YAG systems, the Pulse Delay is the time from
the start of a pulse sequence until the firing of Q
Switch 1.
Delta T (µs) Enter a value for pulse separation. The pulse separation
(dT) is a key parameter for matching the PIV system to
the flow velocity. Choose a pulse separation value so
that the particle image displacements are optimized,
typically less than 16-pixel displacement for a 64-pixel
spot.
See the “Rules of Thumb,” in the PIV Reference Manual
for more information on how to select a pulse separation
value that would lead to good measurements.
Pulse separation in YAG lasers is the time between the
firing of Q Switch 1 and Q-Switch 2.
Note: An identical version of this parameter can be
found on the main capture page just below the
Timing Setup button. Changing the value there has the
exact same effect as changing the value in the timing
diagram, and changing one will always update the other.
PIV Exposure (µs) Enter the time that the first frame on your PIV camera(s)
is open. Refer to your camera operations manual for
specific details.
PLIF Exposure Enter the time that the shutter of your PLIF camera is
open. Refer to your camera operations manual for
specific details.
PLIF Camera Delay Specify an appropriate delay time, in microseconds, for
the PLIF camera. This value is required so that the laser
pulses are situated within the camera exposure.
Num PLIF Laser Indicates the number of laser pulses within a single PLIF
Pulses camera exposure.
Num. PIV Captures Indicates the number of PIV images that will be captured
per PLIF Capture for each PLIF image being captured.

Option Description
Enable: E
Check this boxxto make the synchronizer wait for an
external trigger input before starting a pulse sequence.
t
Leave unchecked to have the synchronizer start the next
e
pulse sequence according to the pulse repetition rate
r earlier in this dialog box.
that was specified
Trigger Delay n(ms): Enter a value for delaying the
camera trigger.aThis value indicates the time from the
start of a pulsel sequence until the camera is triggered.
This parameter is most useful in experiments that use an
external triggerTsuch as rotating machinery with a once-
per-revolution mechanism.
r By adjusting this value,
measurements are made at specific rotation angles.
i
Trigger Timeout (ms): Enter a value for the duration of
g
time that the synchronizer must wait to receive an
external triggerginput. If no external trigger input is
received in theespecified time duration, a timeout
message appears.r
Laser Pulses / Trigger: Specify the number of laser
pulses per external trigger.
Viewing and Adjusting Timing Parameters

Insight software displays the timing settings for the synchronizer, laser and
the camera in a graphical form. Viewing this display you can quickly assess
if all the component timings are adjusted so that when the synchronizer
triggers, the laser is pulsed and the pulses occur in between the two
camera frames.
To View the Timing If the dialog box is collapsed, click on the

Diagram Capture Timing Setup dialog box. The box expands
and the timings you selected for the different
components are displayed graphically.
To Modify the Data Right-click the graph and select the various
Display options. Refer to the online help for specific
information on each of these options.
To Customize the Double-click on the graph. The Timing Diagram
Timing Diagram Customization dialog box opens. Use the options
provided to customize the display. Refer to the
online help for specific information on each of these
options.
To View the Online Right-click the graph and select Help.
Help for the Timing
Diagram

Selecting the Exposure Mode
From the Capture tab, click Exposure selection box and select one of the
following capture mode options.
Option Description
Free Select this to have the frame grabber in the computer
capture and display images as fast as it can, depending on
the camera frame rate.
Synchronized Select this to have the synchronizer trigger control the
camera shutter and the laser pulsing according to the values
set in the Timing Setup dialog box.
Selecting the Capture Mode

From the Capture tab, click Capture selection box and select one of the
following capture mode options.
Option Description
Single Select to acquire images one capture at a time.
Continuous Select to acquire images and display them continuously.
Images are displayed and refreshed at the maximum camera
acquisition rate. When each new frame is acquired, it is
written over the previous frame.
Use this mode initially, when you are setting up the
experiment, to focus the camera and display the images on
the screen.
Note: Continuous mode must be stopped before processing.

Option Description
Sequence Select this to acquire a sequence of captures. The number of
captures is controlled by the values specified in the
Sequence Setup dialog box (see below).
The maximum number of images that can be captured
depends on the camera type, available storage space, and
the memory option you selected. If you exceed the limitation
of your computer memory, an error message is displayed.
To specify Sequence mode values:
A. Click . The Sequence Setup dialog box opens and
is shown in Figure 7-2.
B. Specify the following:
Number of Captures: Specify the total number of files
to be captured.
Start Number: Enter the starting file number. The Start
Number increments automatically based on the last file
number.
Image Save Mode: Select how the files should be
saved. Options include:
 Save Images to Disk
Saves images to the Experiment folder in the
location specified on the computer hard drive.
 Display While Capturing
Displays images while capturing. When this option
is turned off, the frame rate capture performance is
increased.
 Save Images to RAM
Saves images temporarily to RAM. These images
are discarded and not saved when Insight program
is closed.
Note: Insight software displays the estimated disk
space or number of images that can be saved to disk or
RAM.
 Enable Software Trigger
If enabled, the Software Trigger button becomes
active in the control panel (see figure below).
 Post Trigger Captures
The number of files that will be captured after the
Software Trigger button is clicked. In the figure
below, when the Software Trigger button is
clicked, the next 300 captures will be saved to RAM
as are the 1,700 captures prior to the click.
C. Click OK.

Figure 7-2
Sequence Setup dialog on the right. Note the Software Trigger button (currently
inactive) under Capture.
Selecting Laser Power and Pulse Energy

If you are using an Nd:YAG laser in your experiment, depending on the
task you are performing, control the laser power levels by using either the
preset values for the energy per pulse or by changing it.
The energy per pulse for the laser power you select is controlled with the
Q-Switch delay values which are entered in the Laser Energy Setup box.
Edit the Q-Switch delay value to define High, Medium, and Low power for
your laser.
To Check or Set the Q-Switch Delay for Each

Power Level
1. Click on on the Capture tab control panel.
2. The Laser Energy Setup box opens with the Q-Switch Delay/Pulse
Energy default values for High, Medium and Low laser power settings
for the two (A and B) lasers.
3. Change or accept the values using the up and down arrows, or by
entering the appropriate values in the boxes.
4. Click OK.

To Select Laser Power Level
In the Laser A and Laser B boxes on the Capture tab control panel, enter
the desired laser power level. Values include:
Setting Description
Low Sets the power so that the laser is just producing a consistent
green beam. Use Low Power during alignment.
Medium Medium can be used for alignment that requires a little more power
than Low. Could also be used to make measurements in
experiments where full power is not required.
High Sets to a level that gives the maximum laser pulse energy.
Off Sets the Q-Switch Delay value to zero. The flashlamp still fires but
the laser pulse energy is zero.
To Turn the Laser On or Off

Occasionally, during the course of an experiment, you may want to stop
and restart the laser. The following two buttons are available:
Button Description
The laser on button on the Capture control panel starts the laser. It
starts the laser pulsing and causes the external device, camera,
frame grabber, to operate as selected (the cameras capture
images, but will not be retrieved from the frame grabber unless
"Capture" is clicked). The system continues to run until the Laser
Off button is pressed, the number of frames is reached, or an error
occurs.
DANG ER
Wear laser safety goggles when laser is on.
The laser off button on the Capture control panel stops the
synchronizer sequence and turns off the laser and all triggers.
Note: With Nd:YAG systems, clicking the Laser off button once
turns the laser Q-Switch off so that the laser does not pulse but the
laser flashlamps are running at the flashlamp frequency. This keeps
the laser at the operating temperature. Clicking the Laser Off button
twice turns off both the flashlamps and the Q-Switch.
Capturing Calibration Images

Insight software provides options to capture images for calibration. Refer to
“Step C: Recording Calibration Images” and “2-D Calibration and
Perspective Calibration” for detailed information.

Starting and Stopping Image Captures
To Start Capturing Images
1. Select the Capture tab and make sure all the capture parameters are
selected. Alternately, load saved capture settings as described in,
“Saving, Loading and Deleting Image Capture Settings.”
2. Select a run where the image captures will be saved; right-click and
make it current.
3. (Optional) Select the Process After Capture check box if you would
like the images to be processed right after they are captured.
Note: Processing parameters must be selected on the
processing tab.
4. Click Capture. The captured images are stored in the computer RAM
or disk and displayed in the display panel. See “Viewing, Enhancing,
and Displaying Image Files” for details. Also see “Saving Image
Captures.”
To Stop Capturing Images

Click Stop. Image capturing is stopped. The captured images are held in
the computer RAM until they are saved in the current run folder. See
“Saving Image Captures.”
Saving, Loading and Deleting Image

Capture Settings
Insight software allows you to save image capture settings so that they can
be loaded and used for other runs or for other experiments. Alternately,
these settings can be deleted when they are no longer needed.
To Save Capture Settings

1. Select the Capture tab, and make sure all the capture parameters are
selected as desired.
2. Click .
3. In the Save dialog box, enter the name of Capture Setup file.
4. Click OK. The saved file appears in the scroll-down box under Saved
Capture Setups.

To Load Capture Settings
1. In the Capture tab, select the desired capture settings file from the
scroll down box under Saved Capture Setups.
2. Click . The capture settings are loaded.
To Delete Previously-loaded Capture Settings

2. In the Capture tab, select the capture settings file you want to delete
from the scroll down box under Saved Capture Setups.
3. Click . The capture settings file is deleted.
Saving Image Captures

Insight software saves captured images in RAM unless you explicitly save
them on the hard drive of your computer.
To Save Captured Images to the Hard Drive

Click . The images are saved to the current
run in the current experiment folder.

CHAPTER 8
Processing Images
After acquiring images for your application, you can process the images using
the powerful algorithms provided by the Insight™ software. This chapter
describes how to use the processing tab to set up processing for your
application. It also introduces how to setup image preprocessing using the
easy-to-use visual processing pipeline editor, and the various image
preprocessors offered by Insight software. The use of Size Shape Analysis is
also introduced in this chapter.
Processing Tab
The processing tab is the central place for setting up the processing for all
global imaging applications supported by Insight software.
In the processing tab, you can:
Select Application
Following is a list of applications supported by the processing tab. The actual
availability is determined by your software license.
PIV Application 2-D Particle Image Velocimetry where 2-D

vector fields are computed from the particle
images.
Stereo PIV Application Stereo-PIV where 3-D vector fields can be
computed from the particle images of two
camera views.
PLIF Application Planar Laser Induced Fluorescence where
scalar data is computed from the fluorescence
images.
GSV Application Global Sizing Velocimetry where droplet size is
measured from fringe patterns in out-of-focus
images.
SSA Application Size Shape Analysis where particle size and
shape are measured from in-focus images.
Super-resolution Particle Super-resolution particle velocimetry is a
Velocimetry Application particle-tracking based analysis method for PIV
images.
Stereo Automapping Automapping for stereo PIV is used to correct
the misalignment error between the calibration
target and the laser sheet.
8-1
Select Spatial Calibration
2-D Spatial Calibration 2-D Spatial Calibration is needed to convert
length unit from pixels to millimeter. It is optional
in most applications, but is required in GSV
Application.
Perspective Calibration Perspective calibration is required when the
measurement is performed with multiple
cameras such as stereo-PIV, simultaneous
PIV/PLIF, and dual-camera PLIF.
Select Region of Interest

A region of interest is a rectangular region that is specified either numerically or
by clicking and dragging directly on the image(s) to be processed. When used,
only the image area within the region of interest is affected by preprocessing or
processing. More details can be found in the section, “Using a Region of
Interest (ROI)”.
Select Processing Mask

Insight software allows you to define and select any shape in the analysis
region and study flow dynamics in two distinctly separate flow areas. This
feature also helps in saving time and system resources by eliminating areas in
your flows that are not of interest to you and need not be processed and
analyzed. Insight software also allows you to save defined masks so they can
be applied to other images either manually or through a macro. In Insight, there
are two types of masks: static and dynamic. As implied, a static mask is
stationary when applied to the analysis region. A dynamic mask, however,
moves to cover a particular flow feature. Static and dynamic masks can be used
simultaneously and each can be assigned a velocity.
The following sections describe how to:

 Define a static processing mask
 Define a dynamic processing mask
 Assign a velocity to static and/or dynamic masks
 Apply a processing mask
 Delete a processing mask
Select Image Preprocessing

To perform image filtering, image binning, image dewarping, and background
subtraction etc., with image preprocessing, see “Image Preprocessors” for more
information.

Select Processing and Data Post-Processing
Each application has its own unique image processing algorithm. For example,
PIV application performs cross-correlation analysis of the input image pairs;
GSV application performs unique pattern analysis to extract droplet size
information.
Some applications also involve data post-processing steps. For example, PIV
application can have vector validation as post-processing; PLIF application can
have scalar data export as post-processing.
Please refer to the appropriate application chapter on how to setup image

processing and data post-processing.
View Processing Progress

There are two progress bars in the processing tab. The top bar shows the
progress on the current image; and the bottom bar shows the progress on the
entire list of images. There is also a text bar showing the current processing
status. Enabling Turbo Mode will substantially decrease processing times. See
the Display While Processing – “Turbo Mode” section for additional details.
Defining a Static Processing Mask

To define a static processing mask:
1. Select an image(s) from the Exp.
Tree, for which you want to
define a maskDrag and drop it
onto the display panel.
Figure 8-1
2. Go to the Processing Mask
Processing Mask Selection
selection of the processing tab.
3. Select (Create New…) to create a new processing mask (which contains
both static and dynamic mask settings), or select a saved processing mask
to edit.
Processing Images 8-3

4. Click Setup button, the Processing Mask Setup dialog opens (Figure 8-2).
Figure 8-2
Static Masking Setup Dialog
5. Make sure that the Static tab is selected and select Enable.
6. If the opened image has multiple camera views, select the active camera in
the camera drop-down list.
7. Check “Show only area to be processed” to hide the non-processing areas.
8. Click the Invert button to toggle the processing or non-processing area. The
area inside the mask is excluded in the processing. By clicking Invert, the
area outside the mask will be excluded instead.
9. Click Apply Frames to apply current settings to all cameras.
10. Click Generate to create a static mask from the dynamic mask settings.
This can be particularly useful if your static mask has a complex shape that
you don’t want to draw. To Generate a static mask, enable dynamic
masking, following the directions in Creating Dynamic Masks. Once you
obtain a dynamic mask to your liking, disable dynamic masking, return to
the Static tab and click Generate.

11. The following mask shapes are available:
Rectangle Click and drag the mouse in the image window to define a
rectangular mask. Enter the values in the edit boxes to
define the mask. Remember to click Apply button to apply
the new values to the image.
Ellipse Define an elliptical mask is similar to a rectangular mask.
The bounding rectangle of the ellipse is defined.
Polygon Click Start Draw button then use mouse click in the image
to define the vertices of the polygon. To put the vertices on
the edge of the image, click outside the image area. Click
Done button and the last vertex is automatically connected
to the first vertex to close the polygon. Individual vertices
can be modified/specified using the X and Y values. Use the
Point dropdown menu to select an individual vertex to
modify/specify.
Note: To add an additional mask to your current mask, define it while

holding the Shift key.
12. To apply a velocity to the static mask, click Set mask velocity, and enter
the prescribed velocity. Remember to click Apply button to apply the
velocity to the mask.
Defining a Dynamic Processing Mask

Because dynamic masks inherently move, a new mask is created for each
image pair. Mask creation occurs during processing—if no preprocessing,
processing, or post processing is selected, only the dynamic masks are created.
The following sections detail how to define/modify dynamic masks:
 Creating Dynamic Masks
 Altering an Individual (for a single image pair) Dynamic Mask
Creating Dynamic Masks

To define a dynamic processing mask:
define a mask. Drag and drop it
2. Go to the Processing Mask Figure 8-3
Processing Mask Selection
selection of the processing tab.
3. Select (Create New…) to create a new processing mask (which contains
both static and dynamic mask settings), or select a saved processing mask
to edit.
4. Click Setup button, the Processing Mask Setup dialog opens.

5. Select the Dynamic tab and Enable dynamic masking (Figure 8-4).
Figure 8-4
Dynamic Masking Setup
Dialog
6. Choose the appropriate Camera from the dropdown menu. If you want to
use the same dynamic masking settings for both, click All Frames button.
7. To use dynamic masks that have already been created, select Use
Existing Masks.
8. Select Draw Masks On Images During Processing to create a processing
image that is saved in the Analysis folder. It is recommended to have this
option enabled to preserve the correlation plane near mask borders. This
new image will be saved in the Analysis folder and will have an intensity
value painted over the masked region. The intensity value is calculated by
taking the average intensity from all unmasked pixels.

9. The Image Conditioning box contains three tools that are applied only to
help define a dynamic mask. That is, these tools are not meant to pre-
process images. See Image Preprocessors for more information.
a. Image Filtering drop-down
 None – No image filtering is used to define a dynamic mask.
 Local Median –A median filter is applied to the image.
 Local Median Subtraction – A median filter is applied and then
subtracted from the original image. This is the default value
because it commonly removes objects from images while still
retaining particles.
b. Intensity Reduction
The default Size is 30 and is Enabled by default. This tool will reduce
and smooth intensities within a box with the specified Size.
c. Intensity Restoration
The default Size is 50 and is Enabled by default. This tool will increase
and smooth intensities within a box with the specified Size.
10. The Image Masking box contains three tools that are used to help define a
dynamic mask:
a. Thresholding
 Any pixel with an intensity below the specified Value becomes part
of the dynamic mask.
 The button Invert is a convenient tool that when clicked, toggles
regions between masked and unmasked regions.
b. Mask Padding
 This value will pad, or enlarge, the current mask with by the
specified number of pixels. This feature is particularly useful for
objects in the flow that have a strong laser reflection on its edges.
c. View
 Select Grayscale to see the effects of the Image Conditioning
and Image Masking tools. Select Mask to see the actual dynamic
mask (after thresholding) that will be used during processing.
 To apply a velocity to the static mask, click Set mask velocity, and
enter the prescribed velocity. Remember to click Apply button to
apply the velocity to the mask.
11. To apply a velocity to the dynamic mask, enable Set mask velocity and
enter the prescribed velocity. Remember to click Apply button to apply the
velocity to the mask. Dynamic mask velocities are separate from static
mask velocities, and each can be used in conjunction with another.

12. To view the mask settings applied to other images in the image list, simply
progress to a different image using the navigation arrows in the display
panel, highlighted in red in Figure 8-5.
Figure 8-5
Image Navigation Toolbar
Altering an Individual Mask

After creating a set of masks, if you find that the dynamic mask settings did not
create an accurate mask for a particular image (or a smaller set of images),
simply select that image (or set of images) in the Exp. Tree and re-drag and
drop into the display panel. Then, again follow the steps in Creating Dynamic
Masks with different dynamic masking settings. Reprocess so that new masks
are created for that image (or set of images).
Insight 4G software saves all dynamic masks as *.TIF images, which are stored
for the particular Run in the Masks folder. Using a photo editor software, you
can recolor the *.TIF image as you want, save, and have Insight 4G software
use that mask during processing by enabling Use Existing Masks.

Apply a Processing Mask
After you define the desired
processing mask, use it in the
processing by simply selecting it in
the processing tab.
Note: The processing mask does Figure 8-6
not apply to image Processing Mask Selection
preprocessing.
Deleting a Processing Mask

To delete a saved processing mask:
1. Open the Settings folder under current Experiment from the Exp Tree tab.
2. Select the processing mask file from the list, and press the Delete key to
delete the file. You may also use right mouse click to bring the menu with
the delete option.
Using a Region of Interest (ROI)

An ROI is defined directly in the image window by mouse click and drag. If an
ROI is defined on the currently opened image(s), only the selected area is
processed in both image preprocessing and processing. Similar to processing
masks, regions of interest can be saved as a file and reused for different
images.
To define an ROI:
define an ROI. Drag and drop it
2. Go to the Region of Interest Figure 8-7
selection of the processing tab. Region of Interest Selection
3. Select (Create New…) to create a new ROI, or select a previously saved

ROI to edit.

4. Click Setup button, the Define Processing Region dialog opens
(Figure 8-8).
Figure 8-8
Define Processing Region
Dialog
5. Click directly on the image window or manually enter an X (horizontal) or Y

(vertical) starting value (in pixels) and specify the Height and Width (pixels)
of the ROI that you would like to have. Hit Apply Numeric Change to
provide an update to the region on the image window. If more than one
camera is being used, you can specify a separate ROI for each by choosing
either left (L) or right (R) camera. To apply the same ROI on both cameras,
click All Frames.
To Apply an ROI:
1. After you define the desired ROI, use it during processing by simply
selecting it in the processing tab.
To Delete an ROI:
1. Open the Settings folder under current Experiment from the Exp Tree tab.
2. Select the file from the list, and press the Delete key to delete the file. You
may also use right mouse click to bring the menu with the delete option.

Using Image Preprocessing
Image Preprocessing helps improve the raw image quality before application-
specific processing. For some images such as two-phase flow images, image
preprocessing is required in order to separate phases before PIV processing or
size analysis.
To set up Preprocessing Pipeline

1. From the Processing tab, under Preprocessing selection, select Create
New…. After entering a file name for the new setup, the Preprocessing
pipeline box opens. You may also select an existing pipeline and click the
Setup button.
Figure 8-9
Processor Pipeline Editor Screen
2. Select the appropriate image preprocessor from the Processor Module list
in the bottom. The description of selected preprocessor is shown in the
bottom panel. Please see Image Preprocessors for detailed description of
each preprocessors.
3. To add the selected preprocessors to the pipeline, double-click the
processor icon in the list, or use drag-n-drop or the Add button in the top
toolbar.
4. To setup individual preprocessors, double-click the icon in the pipeline, or
use the Setup button in the top toolbar.
5. To remove a preprocessor from the pipeline, select the preprocessor, and
click Remove button in the top toolbar or use the Delete key.
6. Use drag-n-drop to change the order of preprocessors in the pipeline if
needed.
7. To rename the processor in the pipeline, click the name text underneath the
icon.

8. Select Shared Across Cameras option if the defined pipeline is going to be
used by all available cameras. Select Shared Across Frames if the defined
pipeline is going to be used by both Frame A and Frame B. These two
buttons are enabled automatically when there are multiple cameras and/or
multiple frames for the selected application.
9. Click OK to save the current configuration and close the pipeline editor
window.
To Apply a Preprocessing Pipeline

1. Open the images files.
2. Select the desired pipeline configuration file in the drop-down list box in the
Preprocessing section.
3. Click Start. The selected preprocessing pipeline is applied to the images.
Note: When processing and post-processing are also selected, image
preprocessing can run together with the subsequent processing
steps through one click of the Start button. The result of image
preprocessing is always saved.
Image Preprocessors
The following image preprocessors are available to all applications.
Image calculator Perform arithmetic operation between two images or

between an image and a constant; perform image
grayscale inversion and image masking; perform image
rotation and flip.
Image binning Perform image binning to reduce the size of images.
Image filter Perform linear filtering such as Local Mean, Gaussian,
Laplacian, Laplacian or Gaussian; perform non-linear
filtering such as Local Median and Local Range.
Image generator Generate an image from a list of images using average
intensity, minimum intensity or maximum intensity.
The following image preprocessors are available only to certain applications:
Image dewarping Dewarp the images from camera plane to the light
sheet plane using perspective calibration. Only
available in stereo PIV, stereo Automapping and PLIF
applications.
Image deformation Deform the images according to the local velocity data
obtained in PIV. Only available in PIV application.

Image Calculator
Figure 8-10
Image Calculator
Setup
Select an Operation
Arithmetic Addition, subtraction, multiplication, and division
operations between two images, or between an image and a
constant.
Masking operation Mask the input image by thresholding using the
constant threshold or by thresholding based on a mask
image. The mask image can be a static image given by
its file path, or a dynamic image with the same capture
name as the input image with the given processing
number string, or the raw image with the same capture
name as the input image.
Grayscale Invert the grayscale so that the white pixels become
inversion black and vice versa.
Rotation Rotate the image by positive (counter-clockwise) or
negative 90 degrees.
Flip Flip the image horizontally or vertically.
Reverse Order
By default, the input image is the first operand, but the order can be reversed
which may be useful in subtraction and multiplication. For example, 12-bit
image grayscale inversion can be done by subtraction between the input image
and 4096 using the reverse order.

Image Binning
The resultant image size is the input image size divided by the bin size.
Intensity values of the pixels in each bin are summed and divided by the bin
divisor. By default, the divisor is the square of bin size, resulting in the average
value after binning. But any other number can be used as bin divisor.
Figure 8-11
Image Binning Setup
Image Filter
Figure 8-12
Image Filter Setup
Gaussian Filter Gaussian low-pass filter uses a rotationally symmetric

kernel of size n (filter size) and standard deviation
sigma (parameter).
Laplacian Filter Laplacian filter (3x3) approximates the shape of 2-D
Laplacian operator (2nd order derivatives). The
parameter alpha controls the shape of the Laplacian
and must be in the range 0.0 to 1.0.
Laplacian of Laplacian of Gaussian filter is used for edge detection.
Gaussian Filter
Local Mean Filter Replace each pixel by the mean intensity of its
neighborhood whose size is given by the filter size.
Local Median Filter Replace each pixel by the median intensity of its
neighborhood whose size is given by the filter size.

Local Range Filter Replace each pixel by the intensity range (maximum
minus minimum) of its neighborhood whose size is
given by the filter size.
Subtract Source Subtracts the filtered image from the source image.
When used with a Gaussian or Median filter, this option
is a powerful method in removing background features
in your image.
Image Generator
Image generator calculates the pixel-wise average, minimum, or maximum
intensity of a list of input images. The resultant image is usually used as the
background image that can be subtracted from raw images using Image
Calculator.
Figure 8-13
Image Generator
Image Dewarping
Image Dewarping involves pixel intensity interpolation. Define the interpolation
method and window function in the processor setup dialog.
Figure 8-14
Image Dewarping
Setup

Image Deformation
Image deformation processor is usually applied to frame A and frame B
separately. Make sure that Share Across Frame button is the pipeline editor is
unchecked. A vector field must be computed before using the image
deformation preprocessor to set the deformation distances.
Figure 8-15
Image Deformation
Setup
Image Shift
The two frames are shifted in opposite directions and the total amount of shifts
should equal to the local velocity. There are various ways to meet this
requirement. For example, shifting frame A forward by the full shift is equivalent
to shifting frame A forward by the half shift with subpixel and shifting frame B
backward by the half shift integer.
The image deformation can be computed in four ways:

1. Pipeline A Full Shift with subpixel Forward Pipeline B None
2. Pipeline A None Pipeline B Full Shift with subpixel Backward
3. Pipeline A Half Shift with subpixel Forward Pipeline B Half Shift integer
Backward
4. Pipeline A Half Shift integer Forward Pipeline B Half Shift with subpixel
Backward
Image Interpolation
When subpixel shift is used, intensity interpolation is performed during the
image deformation. Setup the interpolation method and window function in the
processor setup dialog.

Background Image Subtraction
A common image enhancement is background image subtraction. Frequently,
the images have some fixed parts like the flow model, walls, background
scattered light, and variable parts like the particles that are being measured. By
removing the fixed parts of the image, the variable part is more apparent and
the signal to noise ratio of the image is improved.
In Insight 4G software, there are two methods to subtract a background:

1. Subtract the background on a per-image basis.
2. Subtract a generated background from a sequence of images.
The first option subtracts a different background from each image, whereas the
second subtracts a single (common) background from each image. The first
option is described in Subtract Source.
Generated background image subtraction requires a sequence of images that

can be used to generate a single background image. Background image
subtraction is a two-step process: 1) generate the background image; 2)
subtract the background image from each image. Insight is able to perform
these two processing steps in one image preprocessor.
The background image is created using an Image Generator. The background

image can be either the Minimum Image or Average Image. The advantage of
using the minimum image is that none of the particle intensity is lost in
subtraction. The average image will subtract some of the particle intensity, but
generally has a better signal-to-noise ratio. Typically first try using the minimum
image and if the signal-to-noise ratio needs improvement try using the average
image.
The background image generator requires enough images so that particles that
happen to be at the same location in a few images do not remove actual particle
images.
To configure a Background Image Subtraction Preprocessor

1. Select Preprocessor, Create New.
2. Drag an Image Generator into the Preprocessor Pipeline.
3. Double-click on the Image Generator icon to open the Image Generator
Setup dialog.
Figure 8-16
Image Generator
Setup

4. Select the Generator Type Minimum Intensity or Average Intensity.
Close dialog with OK.
5. Click on Generator Output Setup from the Preprocessor Pipeline Editor
main menu.
Figure 8-17
Preprocessor Pipeline Editor, Generated Output Setup
6. The Generated Image Location can be set to the Analysis or Settings
directory.
7. The Generated Image Name can be left as the default “Generated” or type
in your own name such as “MinBackgroundImg”.
8. Click OK.
9. Add an Image Calculator to the pipeline after the Image Generator.
10. Double-click the Image Calculator icon to open the Image Calculator Setup
Dialog.
11. Change the Operator to Subtraction.
12. In Operand, select Image.

13. In the Operand drop-down menu change from Image by path to Passed in
Generator image.
Figure 8-18
Image Calculator Setup Dialog for Background Image Subtraction
14. Close the Image Calculator Setup dialog with OK.
15. Close the Processor Pipeline Editor with OK.
When this Background Subtraction Preprocessor runs on an image list, it will

first generate the background image from all the images in the list, and then
subtract the generated background image from each image in the list.
Background Image Subtraction using Two Preprocessors

Background Image Subtraction can be done using two preprocessors with the
first preprocessor generating the background image and the second subtracting
the background image from each image.
The background image is generated the same as in the single preprocessor

background image subtraction, stopping before the Image Calculator is added
to the pipeline.
In the second preprocessor add an Image Calculator to the pipeline.

1. Double-click the Image Calculator to open the Image Calculator Dialog.
2. Select Operation Subtraction.
3. Select Operand Image.
4. In the Operand drop-down menu select Image by path.

5. Click the Browse button. Navigate to the background image generated in
the first preprocessor. Select any one of the images A or B L or R. The
Image Calculator matches the camera and frames.
Figure 8-19
Image Calculator Setup for Two Preprocessor Background Image Subtraction
6. Click OK in the Image Calculator Setup dialog.

7. Click OK in the Preprocessor Pipeline editor.

Size Shape Analysis
Size Shape Analysis (SSA) uses image segmentation algorithm to identify
particles in the image and then computes various size and shape quantities of
the identified particle images. It can also track particles between two frames
using a matching by size algorithm to obtain the velocity data.
Setting up Size Shape Analysis Processor

1. Select Size Shape Analysis application in the processing tab.
2. In the Processing group, select (Create New…) in the drop-down list to
create a new SSA processor; or select an existing processor to edit.
3. Click Setup button, the SSA Processor Setup dialog opens (Figure 8-20).
Figure 8-20
SSA Processor
4. Click Particle Sizing Settings to open SSA Particle Sizing Setup dialog.
Please see Setting up Size Shape Analysis Sizing Processor for more
information.
5. Skip Particle Sizing step to use existing particle data of the opened images
for particle matching to get velocity result.
6. Click Particle Matching Settings to open SSA Particle Matching Setup
dialog. Please refer to “Setting GSV Tracking Processor” for more
information.
7. Skip Particle Matching step if there is only one frame available.

Setting up SSA Sizing Processor
1. Click Particle Sizing Settings button in the SSA Processor Setup dialog,
the SSA Sizing Processor Setup dialog opens (Figure 8-21).
Figure 8-21
SSA Processor Sizing

2. Enter the intensity threshold for image binarization. Use Threshold Assistant
to help you find the optimal threshold value. Click the Threshold Assistant
Apply button to see the image in the window when the Threshold Assistant
is opened, and to see the effect of the current threshold setting
(Figure 8-22).
Figure 8-22
Threshold Assistant
3. Select 4-Connected or 8-Connected as the pixel connectivity which is used
by the algorithm to determine whether two pixels are connected. Two
horizontally or vertically adjacent pixels are always connected; two diagonal
pixels are connected with 8-connected but not with 4-connected.
4. Select the size range of particles to be identified in the input images. One use
of the size range in SSA processor is to do phase separation by size for
multiphase flow images.
5. Select the size and shape measurements to be performed by SSA processor.
The results are exported to the particle data file along with the basic
measurement such as position, diameter, and intensity.
Note: If 2-D spatial calibration is selected in the processing tab, some size
and shape measurements such as Area, Perimeter, and Feret’s
diameter will be exported in the micrometer unit. Others are still
exported in the pixel unit.


CHAPTER 9
2-D Calibration
and Perspective
Calibration
Insight™ software includes the software module Calibration that is used to

spatially calibrate the camera images. The calibration is a map between
camera pixels and world coordinates. The calibration is used in
measurement to convert image displacement into world coordinates. It is
also used in stereo PIV to convert the two 2-D particle image
displacements fields from the left and right cameras into a 3-D velocity
field. The calibration can be used to dewarp images so that they have
constant magnification (constant pixel spacing in world coordinates).
Dewarped images are used with single camera off-axis systems to
measure the velocity field in world coordinates, to remove the perspective
distortion in images, and to match images from multiple cameras pixel for
pixel. Image dewarping is also used in Stereo AutoMapping to optimize the
perspective calibration. The following gives an overview of the 2-D
calibration process and the perspective calibration process.
2-D Calibration Process

Two-dimensional calibration allows you to enter a calibration factor to
compute the flow velocity using meters per second.
Insight software uses the following formula to provide the conversion:
Velocity = (pixel displacement * (mm/pixel)) / dt (in msecs)
If you don’t select a calibration file, Insight software calculates velocity
vector magnitude in pixels (pixels of the camera).
9-1
To Perform 2-D Calibration
1. Open the calibration image.
2. From the Calibration selection of the Processing tab, select Create
New to create a new calibration (or select an existing calibration to edit
it). Click Setup button, the 2D Spatial Calibration dialog opens
(Figure 9-1).
Figure 9-1
2-D Spatial
Calibration
3. Select one of the default origin locations, or manually define the origin
in pixels. Note that the origin can be outside the image; therefore,
negative values and values greater than the image size are allowed.
4. Enter or compute calibration factors using any one of the following
three methods:
 Manually enter Field of View values
 Measure Millimeters per Pixel (mm/pixel) in the image.
 Manually enter Micrometers per Pixel (µm/pixel) values
Field of View
Field of View is the entire image. Insight software uses the mm/pixel
value in the conversion formula. To use this:
A. Place a scale/ruler in the flow.
B. Capture a calibration image. See "Capturing Calibration Images"
for details.

C. Using the scale, enter either the Horizontal or Vertical Field of View
value.
Horizontal (mm) = X mm/pixel  No. of x pixels in the image

Vertical (mm) = Y mm/pixel  No. of y pixels in the image.
These values are all related; editing one causes the other two to
be recomputed. For example:
You have a 640  480 pixel image. Using a ruler you measure the
horizontal field of view in the flow as 50 mm; you enter 50 in the
Horizontal (mm) box.
mm/pixel = 50 mm/640 pixel = 0.078125mm/pixel=78.125m/pixel
Vertical Camera Field of View = 480 pixels * 0.078125 mm/pixel =
37.5 mm
Measured
If you would like Insight software to calculate the millimeters per pixel
value:
A. Place an object of known size, such as a flow model or a
scale/ruler in the flow.
B. Type in the object size in millimeters in the Object Size in the mm
box.
C. Click Measure.
D. Place the cursor on the start of the object. Click the left mouse
button and drag to the end of the object. Release the mouse
button. Based on the pixels measured using the cursor, the
calibration factor of xx mm/pixel is calculated.
Manual
If you know either the x or the y µm/pixel value, directly enter that value
in the appropriate box and click the Apply button.
5. Click OK. The calibration file is automatically saved.
2-D Calibration and Perspective Calibration 9-3

Perspective Calibration Process
Following is an overview of the steps involved in a perspective calibration.
The perspective calibration is usually performed for stereo PIV
arrangement or system using a single camera with off-axis viewing to the
flow field.
Step A Align the cameras with the calibration target using Insight
to view the images (multiple camera systems only).
Step B Fine tune the Scheimpflüg angle and focus with flow
images and record flow images using Insight software.
Step C Record images of the calibration target using Insight
software.
Step D Process these calibration images to create a new
Perspective Calibration file.
Step E AutoMap to correct calibration target alignment errors
using Insight software (Stereo PIV only).
Perspective Calibration Overview

Perspective Calibration analyzes the calibration target images that were
captured with Insight software. The image analysis finds the location of
each calibration marker point in the sequence of image frames and
matches the image (X. Y) location to the target marker (X, Y, Z) location in
the fluid. This set of calibration points is used to create a calibration
mapping function.

Calibration Targets
The calibration or mapping process uses a calibration target, which is a
rectangular grid of marker points with known (X, Y, Z) locations. Two types
of targets may be used single plane or dual plane.
Figure 9-2
One-Plane Target
A single image of a single plane target may be used for one camera
systems or multiple camera PLIF to create a XY (no Z) calibration for
dewarping. A dual-plane target may also be used for these applications.
For stereo PIV the single plane target is mounted on a traverse and images
from each camera are captured at multiple Z to create a XYZ calibration. A
traversed single plane calibration target is generally more accurate than a
dual plane target. With the traversed single plane target twice the number
of calibration points per plane are used than a dual plane target with the
same grid spacing. Using 5 to 9 calibration planes allows a larger depth
and higher order Z polynomials to be used. The single plane target is
typically used to calibrate situations with high distorted medium between
the cameras and the flow region. For example, if the flow measurement is
in a water tunnel and the wall of the tunnel is curved in shape, it would be
more desirable to use a single-plane target to provide the calibration
relationship. It is because the multiple planes (up to 9 planes) in the
Z direction can provide higher order Z polynomials for the calibration
curve fit.

Figure 9-3
Four-Plane Calibration
Target
The two-plane target has calibration marker points that alternate between
two depths. This allows a 3-D calibration to be performed without having to
traverse the target. Only a single left and right image is used with the two-
plane target. The advantage of the dual-plane target is that it does not
have to be traversed. In many experiments access to the flow area is
limited and traversing a target impractical. The two-plane target allows
measurements in these more limited access experiments.
In experiments where the cameras are on both sides of the light sheet a
four-plane target is used. It is the same as the two-plane target with a
second calibration dot grid on the back.
The target is placed in the laser light sheet plane. Automapping is used to
correct the target-light sheet misalignment in a stereo PIV system. The
Automapping function helps to re-calculate the calibration polynomials and
to correct for the mis-alignment of the target.
Details on how to set up this grid are provided in the PIV Systems Manual.

Stereo PIV Camera Location
The cameras may either be located on the same side of the light sheet
(Figure 9-4) or on both sides of the light sheet (Figure 9-5). The choice in
configurations is largely dependent on your experiment and flow model.
Having both cameras on the same side of the light sheet, as shown in the
following figure, requires less physical and optical access to the
experiment.
Figure 9-4
Stereo PIV Camera Location (one side)
In the following figure, the laser light sheet passes between the cameras.
The left camera has a negative Z position, the right camera has a positive
Z position.
The main advantage of having the cameras on both sides of the light sheet
is that both cameras may be in forward scatter where the light intensity is
higher. Also the camera fields of view can be closely matched with the
cameras on both sides of the light sheet. In the following figure both
cameras are on the same side of the laser light sheet, and have positive
Z positions.

Figure 9-5
Stereo PIV Camera Location (two side)
Step A: Overlapping the Camera Fields of

View with the Calibration Target
Before calibrating, the cameras must be aligned and the Scheimpflüg angle
set. The camera alignment is done by first setting the camera angles using
a target, then setting the focus and the Scheimpflüg angle looking at
particles in the laser light sheet. Although the focus and the Scheimpflüg
angle can be set by looking at the target, looking at particle images is more
precise. In this step the camera angles are set and the Scheimpflüg angle
is approximately set. The final Scheimpflüg angle and focusing are done in
step B.
To Align the Cameras

1. Align the camera using the calibration target or any other target where
you can relate the image to a physical location. Because the
magnification varies across the field of view, the best overlap is not
when the image centers are viewing the same point. Instead, check the
left and right edge of the images. This is most useful when both the
cameras are on the same side of the light sheet configuration. Use the
image frame Tools | CrossHair option to help in overlapping the
image.

2. Focus the camera.
3. Loosen the Scheimpflüg mount lens angle adjustment.
4. Rotate the cameras and lenses to overlap the camera fields of view.
Refocus if necessary.
5. Tighten the lens angle adjustment on the Scheimpflüg mount.
6. Loosen the Scheimpflüg angle adjustment.
7. Tilt the camera body so the calibration dots are in focus from edge to
edge in image. Refocus if necessary. This is only a rough Scheimpflüg
angle adjustment. The dots are too big to determine precisely when the
Scheimpflüg angle or focus is correct.
8. Tighten the Scheimpflüg angle adjustment.
9. Verify that the camera fields of view are well overlapped. If not repeat
the lens angle adjustment.
10. Tighten the Scheimpflüg mount lens angle adjustment. Do not make
any changes to the lens angles after this step. If the lens angle does
change the calibration will be invalid and have to be done again.
Step B: Setting Scheimpflüg Angle and

Focus with Particle Images, Record Flow
Images for AutoMapping
The calibration target does not have enough precision to accurately set the
Scheimpflüg angle and focus. The dots on the calibration target are too
large to determine precisely when they are in good focus. And unless the
target is very accurately positioned in the light sheet plane focusing on the
target does not assure the particle images will be in focus.
1. Remove the target.
2. Image the flow with the laser light sheet. With particle images you can
discern whether they are in focus or out of focus better with a
calibration target. The particle images also have a high resolution for
setting the Scheimpflüg angle. Using a large lens aperture setting
(small F/#) minimizes the depth of field and enables increased
adjustment precision.
3. From the menu bar of the image file in the display panel, select Size |
Zoom In. For the most accurate view of the Scheimpflüg angle setting,
zoom into 1:1 magnification or higher and pan from left to right.
4. Loosen the Scheimpflüg angle adjustment.
5. Adjust the Scheimpflüg angle and focus until the particles are in good
focus across the image.
6. Lock down the Scheimpflüg angle and do not refocus the camera after
this step.

7. Repeat the Scheimpflüg angle adjustment for the other camera.
8. Record a sequence of particle images and save to disk. These images
can be used for AutoMapping and calibration verification while the
target is still in position in the next step. In these images the flow is not
important. AutoMapping only uses the A frames so particle
displacement is not important. Save at least 10 images, more if the
particle concentration is low. These AutoMapping images may be
deleted after the calibration has been verified.
Step C: Recording Calibration Images

In this step you capture and save a set of images of the calibration target
that can be analyzed and used as data points for the calibration file. Refer
to the PIV Installation Manual for details on hardware setup. Perform one
of the following two sets of procedures depending on which calibration tool
you are using:
 Single-plane target mounted on a traverse
 Two-plane or four-plane target.
Single-Plane Target Mounted on a Traverse

If you are using a single plane target mounted to a traverse, follow these
steps.
1. Place the target in the laser light sheet using the mirror with slit to align
the target to the laser light sheet. Use a bright light to illuminate the
target, and a large f/# lens setting so that depth of field is high enough
to have the all the target points in focus.
2. Click the Capture tab and make the following selections from the
Capture control panel:
Parameter Value
Exposure Free
Capture Continuous
Note: Be sure the Start Capture button is not activated. Having the
system capturing live video when starting the Capture PIV
Calibration Images or the Capture PLIF Calibration Images
process, can cause distorted images.
3. Select Tools | Capture Perspective Cal. Images from the main
Insight menu. The Perspective Calibration Image selection box
appears.
4. Click Capture. A pair of left and right images are captured and stored
in the experiment’s setting folder. 000001in the filenames is the image
sequence index number. This number is always 0 when the Capture
PIV or Capture PLIF Calibration Images dialog box is opened.

This index number is used later by the Perspective Calibration program
to compute the target Z-axis location. The calibration image sequence
should start with 0 so that the Z-axis positions are computed correctly.
The formula for converting sequence number to target z position is
listed later in this section. It is recommended that the entire calibration
image capture sequence be performed in one Capture PIV or Capture
PLIF Calibration Images session. If for some reason one of the images
in the sequence has a problem, the entire sequence of images should
be captured again. This prevents introducing errors in calibrations
resulting from traversing back to the correct Z-axis location in the
middle of the image sequence. To capture a new sequence move the
traverse to the start position. Close the Capture Calibration Images
selection box and reopen it to reset the sequence count.
z = z Start + ( Sequence Number * z Step )

5. Move the traverse to the back of the light sheet thickness. For
example, to map a 1-mm thick light sheet with five target planes, start
with the target at -0.5 mm and translate it in +0.25 mm increments.
6. Click Capture again on the appropriate Capture Calibration selection
box to save the image at this z location.
7. Traverse forward one step and save left and right camera images.
Continue traversing and saving until images from all calibration planes
are saved.
Two-Plane or Four-Plane Target

If you are using a two-plane or four-plane target, follow these steps.
1. Place the target in the laser light sheet using the mirror with slit to align
the target to the laser light sheet. Use a bright light to illuminate the
target, and a large f/# lens setting so that depth of field is high enough
to have the all the target points in focus.
2. Click the Capture tab and make the following selections from the
Capture control panel.
Parameter Value
Exposure Free
Capture Continuous
Note: Be sure the Start Capture button is not activated. Having the
system capturing live video when starting the Capture PIV or
Capture PLIF Calibration Images process, can cause distorted
images.
3. Select Tools | Capture Perspective Cal. Images from the main
Insight menu. The appropriate Capture Calibration Image selection box
appears.
4. Click Capture. A pair of left and right images are captured and stored
in the experiment’s setting folder. With the two-plane target only one
set of calibration images are used.

Step D: Processing the Calibration Images
and Creating a New Calibration File
In this step the calibration images acquired in Step C, are analyzed using
the Perspective Calibration program to create a new calibration file.
Calibration Processing: Overview

The following is an overview of all the steps that are involved in processing
a calibration image. Details on each of these steps are provided in the
following sections.
Until you are familiar with the calibration process, you may want to follow
and perform all these steps. Once you are familiar with the process, some
of these steps may become optional.
In some experiments, all of the parameters can be set so that all calibration
images can be analyzed together. In many experiments, it is easier to
analyze the left camera images with one set of parameters and then the
right calibration images with a different set of parameters.
To Create a New Perspective Calibration:

1. In the Insight Processing Tab select the Application then Calibration
(Create New…) The Perspective Calibration program opens.
2. Click Create New CPT Files in the Perspective Calibration dialog. The
Calibration Image Analysis dialog appears.
3. Select the camera first. Then click Select Calibration Image(s). The
Open TIF Image box opens.
4. Select all the calibration images of the current camera. The Files of
Type filter is automatically set to show the calibration images from only
the current camera in the image list.
5. Enter a Threshold value. Refer to “Setting the Threshold Value” for
details.
6. Click Define Target. See "Defining the Target” later in this section for
information on how to define the target and the Fiducial Marker Based
Grid.
7. Click Analyze One to analyze the current image. Inspect the screen
display to make sure there are boxes around the calibration markers
and a diamond around the fiducial mark in the center.
If there are errors, repeat step 5 and enter new threshold values. Also,
click Options and perform the steps outlined in “Analyze Options
Dialog,” later in this section, to define area of interest, object size, and
grid tolerance.

8. Click Analyze ALL to find the calibration markers in all selected
images. The pixel location of each calibration marker on the calibration
target is stored in a calibration points (*.CPT) file which is created from
each image file. This file has the same name as the image file but with
a .CPT file extension; which is stored in the calibration folder.
9. If some camera calibration images have not been analyzed repeat step
3 to select all the images from the camera. Repeat steps 4 to 8 with
these image files.
10. If the calibration images for all cameras have been analyzed, click Go
back to Main Calibration Window button, or close the Calibration
Image Analysis dialog to go back to the initial Perspective Calibration
dialog.
11. Select a camera in the Camera Calibration group box. Click Edit
Camera to open the Camera Target Calibration dialog.
12. Enter the order of the polynomials you want to use in the resulting
calibration. Click Select CPT Files to select the CPT files you just
created for the current camera. Click OK to return to the previous
dialog.
13. Repeat steps 11 and 12 to define each camera in your system.
14. Press Compute Calibration to create the perspective calibration.
Close the System Calibration dialog unless the calibration images for
all cameras have been analyzed.

Analyzing Calibration Images
In the Calibration Image Analysis dialog, select the camera you want to
process. Click Select Calibration Image(s) button to select the calibration
target images of the current camera.
Figure 9-6
Calibration Image
Analysis
Setting the Threshold Value

When the frame is being analyzed for calibration marker images, it uses
the threshold value to determine if the pixel is “lit” or not. When a pixel is
greater than or equal to the threshold value, it is lit. The Calibration Image
Analysis program joins the lit pixels into objects that identify the location of
the calibration marks in the image.
If the calibration target is not light dots on a dark background press the
Positive button to invert the gray scale values (see “(see “Choosing
Positive or Negative Settings”). Generally, the threshold is used to
separate the marker pixels from the background.

This threshold level is a simple test for identifying the lit pixels. For this to
work well, the images must have a clear intensity separation between the
markers and the background, and this level must work over the entire field
of view. Therefore, during the image capture process, you should adjust
the exposure and lighting to create this intensity difference.
The edges of the images may be excluded from the marker analysis using
the controls in the Options dialog.
1. You may select a different color scheme for displaying the lit and
background pixels if you want. Click Blue/Red | Gray. See “Choosing
Blue/Red | Gray and Color Schemes” for additional information.
2. The histogram display may help you in selecting the correct threshold
level. Click Histogram if you want to view the intensity histogram. See
“Using the Histogram to Set Threshold Level.”
3. To set the threshold value, enter the appropriate values in the
Threshold box. The Threshold value can be set by using one of the
four buttons -10, -1, +1, +10, or by typing the threshold value in
directly.
Defining the Target

The target definition dialog is used to define the fluid space (x, y, z)
coordinates of markers in the calibration points file.
1. On the Insight main menu bar, select Tools | Perspective Calibration.
The Perspective Calibration program opens.
2. Click Image. The Calibration Image Analysis dialog box appears.
3. Click Define Target. The Target Configuration dialog box appears.
Figure 9-7
Target Configuration
4. Select the appropriate Target Surface Plane option:

Option Description
1 Plane, z Traverse Select for a single surface target with traverse.
2 Plane, no Traverse Select for a two-plane target with no traverse

5. Select the appropriate Camera Location option:
Option Description
Same Side of Light Select if camera or cameras are located on
Sheet (+z) same side of the light sheet.
Both Sides of Light Select if camera or cameras are located on both
Sheet (Left –z, Right +z) sides of the light sheet.
6. Click Define Target. Follow the appropriate set of instructions given in

“To Define the Single-Plane Fiducial Marker Grid Values” or “To Define
the Two-Plane Fiducial Mark Grid Values”.
With the fiducial marker-based calibration and the standard calibration

target, the cross in the center of the target is the fiducial mark. It can be
identified as the mark larger than all the other calibration markers. The
fiducial marker is highlighted with a diamond during the display of the
calibration points. With fiducial marker calibration, you enter the (x, y, z)
location of this cross in the first calibration image. The grid is validated
using the fiducial marker as the starting point, with each grid node having
an index relative to the marker point. One advantage of using the fiducial
marker calibration is that you do not have to set up the calibration target
with a specific number of calibration points. The software is able to find all
of the markers and reject points that are too close to the edges of the
image. This method allows for calibration points to leave the valid image
area as the target is traversed. It also allows the left and right cameras to
view slightly different fields of view.
To Define the Single-Plane Fiducial Marker Grid Values

1. Perform steps 1-6 outlined in “Defining the Target.”
2. Enter the appropriate values in Fiducial Mark Based Grid dialog box.
Following are descriptions of the parameters available on the single-
plane grid dialog boxes:

Parameter Description
x & y mm Fiducial Enter (x, y) value in mm for the fiducial mark in the
Mark Coordinates 3-D vector file. For example, if your camera field of
view is 50 mm  50 mm and you leave the fiducial
mark (x, y) set to (0, 0), your vector field will be from
(-25, -25) to (25, 25). If you change the fiducial mark
(x, y) to (25, 25), your vector field range will be from
(0, 0) to (50, 50).
Note: These example numbers are approximate
and assume that the fiducial mark is in the center of
the image, that the two images are well overlapped
and that you are processing the whole images and
your points file covers the entire image.
If you want to have precise coordinates for some

point in the vector field:
a. Create a calibration using approximate
coordinates for the (x, y) fiducial mark.
b. Process a vector field and see what the
coordinates are in the vector field output.
c. Compute the error in the field of view
coordinates.
d. Reprocess the calibration images used adjusted
fiducial mark (x, y) coordinates.
z mm First Target The z-axis location for the first calibration image
Plane files, image sequence number 00. For example, if
the first traverse position is 0.5 mm from the center
of the laser light sheet away from the cameras, enter
the value -0.5 mm. If the first traverse was 0.5 mm
towards the cameras from the center of the laser
light sheet, enter the value 0.5 mm.
x & y Grid Spacing Enter the distance between the horizontal grid points
(mm) in the x grid spacing box and the vertical grid
spacing in the y grid spacing box.
z Grid Spacing mm The z grid spacing is the distance traversed
(can be + or -) between each calibration image capture. If the
traversing moves the target closer to the cameras,
the value is positive. If the traversing moves the
target away from the cameras, enter a negative
value.
Note: The calibration image file sequence number is
used to compute the z-axis position. It is important
that when capturing a sequence of calibration
images that you keep the sequence numbers
correct.

3. Click OK. The Grid Defined box is checked.
Figure 9-8
Two Plane Grid
Note: The grid definition dialog changes depending on which of

these configuration parameters are specified. The grid
definition is saved in a file named GridDef.txt in the settings
folder of the current experiment. If this file is found, the saved
grid definition parameters are filled in the grid definition dialog
box. If a GridDef.txt file is located, it appears in the Target
Configuration dialog.
To Define the Two-Plane Fiducial Marker Grid Values

1. Perform steps 1 to 6 outlined in “Defining the Target.”
2. Enter the appropriate values in the Two-Plane Grid dialog box.
Following are descriptions of the parameters available on the two-

Fiducial Mark Enter (x, y) values in mm for the fiducial mark. For
Location example, if your camera field of view is 50 mm  50
(x, y) mm mm and you leave the fiducial mark (x, y) set to (0, 0)
your vector field will be from (-25, -25) to (25, 25). If
you change the fiducial mark (x, y) to (25, 25), your
vector field range will be from (0, 0) to (50, 50).
Note: These example numbers are approximate and
assume that the fiducial mark is in the center of the
image, that the two images are well overlapped, and
that you are processing the whole images and your
points file covers the entire image.
If you want to have precise coordinates for some

point in the vector field:
a. Create a calibration using approximate
coordinates for the (x, y) fiducial mark.
b. Process a vector field and see what the
coordinates are in the vector field output.
c. Compute the error in the field of view
coordinates.
d. Reprocess the calibration images used to adjust
Fiducial Mark z The z-axis location for the fiducial marker on the
Plane (mm) target.
For example if the two target planes are 1 mm apart
and the fiducial mark is on the raised surface (closest
to the camera), enter +0.5 mm for the fiducial mark z
location and -1.0 for the target z spacing. If the
fiducial mark is on the recessed plane (away from the
camera), enter -0.5 mm for the fiducial mark z
location and +1.0 mm for the z spacing.
x & y Grid Spacing Enter the distance between the horizontal grid points
(mm) in the x grid spacing box and the vertical grid spacing
in the y grid spacing box.
z Grid Spacing mm The z grid spacing is the distance between the two
(can be + or -) target surfaces. (See the z fiducial mark location
entry above for example.)
3. Click OK when you are prompted to save changes.

To Define the Four-Plane Fiducial Marker Grid Values
1. Perform steps 1 to 6 outlined in “Defining the Target.”
2. Enter the appropriate values in the Four-Plane Grid dialog box.
Following are descriptions of the parameters available on the four-
Fiducial Mark Enter (x, y) values in mm for the fiducial mark. For
Location example, if your camera field of view is 50 mm  50 mm
(x, y) mm and you leave the fiducial mark (x, y) set to (0, 0) your
vector field will be from (-25, -25) to (25, 25). If you
change the fiducial mark (x, y) to (25, 25), your vector
field range will be from (0, 0) to (50, 50).
Note: These example numbers are approximate and
assume that the fiducial mark is in the center of the
image, that the two images are well overlapped, and that
you are processing the whole images and your points file
covers the entire image.
If you want to have precise coordinates for some point in

the vector field:
a. Create a calibration using approximate coordinates
for the (x, y) fiducial mark.
b. Process a vector field and see what the coordinates
are in the vector field output.
c. Compute the error in the field of view coordinates.
d. Reprocess the calibration images used to adjust
Fiducial Mark The z-axis location for the fiducial marker on the target.
z Plane (mm) For example if the two target planes are 1 mm apart and
the fiducial mark is on the raised surface (closest to the
camera), enter +0.5 mm for the fiducial mark z location
and -1.0 for the target z spacing. If the fiducial mark is on
the recessed plane (away from the camera), enter -0.5
mm for the fiducial mark z location and +1.0 mm for the z
spacing.
x & y Grid Enter the distance between the horizontal grid points in
Spacing (mm) the x grid spacing box and the vertical grid spacing in the
y grid spacing box.
z Grid Spacing The z grid spacing is the distance between the two target
mm (can be + surfaces. (See the z fiducial mark location entry above
or -) for example.)
2. Click OK when you are prompted to save changes.

Note: When the cameras are on both sides of the light sheet, the
calibration software expects the right camera to be looking at
the front-side of the light sheet and the left camera to be looking
at the back-side of the light sheet. The left camera image x-axis
is opposite the right camera x-axis because it is looking from
behind.

Analyze Options Dialog
Use the Calibration Image Analysis options dialog to:
Define or select an area Use a set of inputs to reduce the area of

of interest to look for interest for the image analysis, setting
calibration marker points different number of pixels to exclude for the
left, right, top and bottom edges.
Define object size By defining an object size, only objects
greater than or equal to the minimum
object size and less than or equal to the
maximum size will pass through the filter
and be used when matching the image
objects to the grid definition.
Define the Grid Tolerance These controls adjust how close calibration
and Connection order points need to be to existing grid
prediction.
Figure 9-9
Calibration Image
Analysis Options
To Specify the Region of Interest and Object Size

3. Click Options. The Calibration Image options dialog box appears.
4. Enter the appropriate values for the parameters described below:

Image Size Pixels Displayed for reference purpose only. The first
number is the image width, the second the image
height in pixels. The left and right search limits
must be between 0 and the image width. The
bottom and top search limits must be between 0
and the image height.
Object Search Area Pixels Enter different number of pixels to exclude for the
left, right, top and bottom edges.
Auto Search Area: Selects the entire image for
image analysis. If you have entered limit values but
Auto Search Area is selected, your input limits are
ignored.
Manual Search Area: Reduces the area of
interest. When this button is checked, the input
area is the image analysis area of interest.
Edge Pixels: Excludes the same number of pixels
from all four edges.
Object Size Limits (Pixels) Enter a value to help filter the objects found in the
image analysis by size. Only objects greater than or
equal to the minimum object size and less than or
equal to the maximum size pass through the filter
and will be used when matching the image objects
to the grid definition.
Grid Connection Order Used when an image has a large amounts of
distortion. Allows the calibration point to be verified
with higher order location predictors.
Grid Tolerance Pixels Used when calibration points are being added to
the grid. A calibration point must be within this
tolerance of the predicted pixel location to be
added to the grid. The tolerance must be loose
enough to allow for the image distortion that is
being measured but tight enough so that it finds
only the correct points.
5. Click OK.
Tools and Troubleshooting

Choosing Positive or Negative Settings
The image analysis is designed to work with bright calibration markers on a
dark background. This is a positive image and you should select the
positive setting. If you are using the TSI-supplied calibration plate, use the
default setting of Positive.
Sometimes it is easier to generate a calibration target that has dark

calibration markers on a bright background, this is a negative image. Use
the Negative button to invert the pixel values in the image and in the image
display.

To Toggle Between Negative and Positive Settings
1. On the Insight main menu bar, click Tools | Perspective Calibration.
3. Click Positive to toggle between the two values.
Choosing Blue/Red | Gray and Color Schemes

After opening a calibration image and before selecting the threshold value,
use the Blue/Red | Gray button to help you select the correct threshold
values. This feature is designed to help you select the correct threshold
values by highlighting the calibration markers and isolating them from the
background.
The Blue/Red color scheme shows pixels above the threshold value as red
and pixels outside the threshold values as a shade of blue. The red pixels
are “lit” and they are joined together to define the calibration markers. This
allows you to visualize which pixels would be included in the marker points
for analysis. If the markers cannot be isolated from the background, you
need to capture the images again. Change the lighting or exposure so that
the contrast is better or the background is more uniform the second time.
The Gray color scheme shows the image as shades of gray. This is the
normal view of the image.
To Select a Color Scheme

1. Click Perspective Calibration from the Insight main menu bar. The
Perspective Calibration program opens.
3. Click either Blue/Red or Gray to select one of the two color schemes.
Using the Histogram to Set Threshold Level

Use the histogram to help you set the correct threshold levels. The
histogram shows the number of pixels at each intensity level. The X-axis is
the histogram bin or pixel intensity level from 0 to 255. The Y-axis is the
number of pixels in the bin. The histogram uses the Positive/Negative
setting (See “Choosing Positive or Negative Settings”) when displaying the
pixel intensities. This means that the histogram intensities are reversed if
negative setting is selected.

To Display the Histogram
3. Click Histogram. The histogram appears.
Figure 9-10
Perspective Calibration
As shown in Figure 9-10, the histogram shows the pixel intensity

distribution in two ways—number of pixels at each intensity level and
cumulative pixels less than or equal to the intensity level. The bins are
drawn using the color for that intensity in the image. Viewing the histogram
with the Red/Blue color scheme can show exactly where the threshold
value lies in the intensity distribution. The number of pixels in a bin can be
read from the counts labels on the right side of the histogram. The ratio of
the number of counts in a bin to the number of counts in the highest bin
can be read from the percentage scale on the left of the histogram. The
magenta line is the cumulative histogram, showing the number of pixels
less than or equal to a pixel value. The cumulative histogram value is read
as a % of total pixels from the left axis.
Although Perspective Calibration is able to read 12-bit Insight software TIF

images, it can only operate on the eight most significant image bits and
shows the scale as 0 to 255. If the actual image maximum pixel value is
4095, the histogram scale is off by a factor of 16. If the grayscale values in
the 12-bit image are all less than 255, the image is scaled 0 to 255 and the
histogram scale is the same as the image scale. To verify the actual
grayscale range of an image, use Insight software and set the Tools | LUT
to Pseudocolor. You will be able to see the pixel value range by the range
of colors in the image.

When the histogram bins are shown with normal scaling, the histogram
may just show a spike at the background intensity level with nothing visible
in the foreground. By changing to a Log scale factor, bins with fewer counts
may show on the graph.
To Change the Y Log Scale Factor in the Histogram

Click Log/Normal. The scale factor cycles from Normal, 2 Decade Log, 3
Decade Log, 4 Decade Log, 5 Decade Log, 6 Decade Log.
Verifying Calibration Markers in the Calibration Points and

Calibrations Files
During the batch processing the calibration markers are highlighted and
boxes drawn around them to show the results of the image analysis. This
highlighted view flashes quickly to verify that the image analysis did, in fact,
find the calibration markers correctly.
To Verify Markers
1. On the Insight Processing Tab with a Stereo PIV Application selected
select a Calibration then press Setup to open the Perspective
Calibration program.
2. Close the System Calibration dialog. Click Image in the main menu of
the Perspective Calibration program. The Calibration Image Analysis
dialog box opens.
3. Click Select Images and open one or more calibration images.
4. Check Show CPT Pt.
Diamonds are drawn around the fiducial point and a box around the other
target points on the same plane as the fiducial mark. If a two-plane target is
being used, the calibration points not on the fiducial plane are shown with
circles.
Note: A calibration points (CPT) file must exist for the calibration markers
to be shown.
By looking at the analysis results you can verify the fiducial mark and the
calibration markers were properly identified, and that noise was not used
as a calibration marker. If a few noise spots are identified as calibration
points, remove them by doing the following:
To Fix Improperly Identified Markers

Rescan image with a different threshold value if too many of the markers
are not properly identified. If a good threshold value cannot be found,
recapture the calibration images with better lighting and exposure settings
so that the calibration markers can be isolated from the background.

It is acceptable to have an image where a few of the calibration markers
were not located. There are several reasons why a calibration point could
be missed:
 If it is too close to the edge of the image, it is eliminated.
 If the threshold value does not isolate the marker pixels from the
background, it is not identified.
 If the marker image is broken up into unconnected pixel groups, all of
the marker groups will be eliminated. If there is a bad pixel in the
calibration marker image, it will be eliminated. If there is a bright spot
like a reflection or bubble in the image between calibration markers,
both the good marker and the noise marker will be removed so that
only good markers remain.
The fiducial marker may not be properly identified if it is not in the central
area of the image, or if it is not the largest object in the image. You may be
able to use the Options… dialog to configure the software to work with
your image.
The calibration is computed using a least-square fit of all calibration points.

If a few marker locations are missing, a good calibration can still be
computed.
Figure 9-11
Figure 9-11 shows the results of an image analysis. The fiducial mark
cross is highlighted with a diamond and the calibration marks are
highlighted with boxes. Two marker dots in the center of the grid are not
highlighted. This is because there are bright pixels between these two
markers. In the grid validation step the local grid spacing for these dots
was found to be outside the tolerance and these points were removed from
the grid. This removed both the extraneous bright pixels and the calibration

markers. This image still has enough good calibration points to be used to
create a calibration file.
Four calibration marks very close to the right edge of the image are visible
in the lower-right corner. These marks were eliminated for being too close
to the image edge.
After a calibration file has been created, you can view a grid of lines from
the fluid mapped into the image with the Show Lines checked. The lines
are located midway between calibration points and mapped into the fiducial
plane. The boxed calibration points should be centered between the lines.
Circled points are not on the fiducial plane and will not be centered
between the lines. Look at the edges of the calibration grid, the mesh
should not become too warped or fold upon itself. If is too warped, using a
lower equation order can improve the calibration—especially at the edges
of the image. If the grid looks good, you may want to try increasing the
calibration equation order for improved accuracy. If you use a higher order
grid, be sure to view it again to verify that it looks okay.
Creating a Calibration File from the CPT Files

After the images have been analyzed, the calibration points are stored in
files with the *.CPT extension. The next step is to create a calibration file
from these calibration points and view the graphical display to verify that
the calibration was done properly and check if:
 All the calibration points appear in a reasonable grid in the camera
fields of view
 All the camera fields of view look like trapezoids in the fluid point of
view.
To Create a Calibration File

1. In Insight software, select
Create New… from the
Spatial Calibration drop-
down list in the processing
tab.
Figure 9-12
Create New Calibration

2. Enter a name for the new calibration file and click OK. The Perspective
Calibration dialog opens (Figure 9-13).
Figure 9-13
Perspective
Calibration
3. The Number of Cameras and the name of the cameras are set by the
Insight |Tools| Hardware Setup dialog.
4. Click Select a Camera from the drop-down list. Click the Edit Camera
button to open the Camera Target Calibration dialog (Figure 9-14).
Figure 9-14
Camera Target Calibration

5. Press the Select CPT Files and select the calibration files to be used
for this camera. With single plane targets with traversing select all of
the *.CPT files for the camera. To change the selected files, select all
of the calibration points files for the calibration. You cannot add or
remove individual files from this list. To select multiple files, hold the
control key down and click on each file, or to select sequential group of
files, hold the <Shift>+<Ctrl> keys then click on the first and last file in
the list.
6. Enter the mapping polynomial equation orders. The first group of
numbers is the x y and z order of the mapping between the camera
and the world. The second set of numbers is the xy and z order
between the dewarped image and the world. The polynomial mapping
equations can be from 1st to 5th order for each axis. Generally, using
high order equations can give better accuracy, but the conversion from
two 2-D to a 3-D vector file will take a little longer. Using equation
orders that are too high may result in mapping functions noise. This
may be a problem when the calibration grid does not fill the image. To
looking at the lines in the image dialog will show if the equations go
wild outside the calibration dot area. The maximum equation order that
you should use is number of calibration points in that axis -1. For
example if you are using 3 z calibration planes, 2 is the maximum z
equation order that you should use.
7. Click OK to return to the Perspective Calibration dialog.
8. Click Compute Calibration button to create the calibration file using
selected CPT files for all the cameras in the system. The calibration file
is located in the settings folder of the current experiment.

Viewing a Calibration File
1. In Insight software, select an existing calibration file from the Spatial
Calibration drop-down list in the processing tab.
2. Click Setup button next to the drop-down list. The Perspective
Calibration dialog opens with the stored information shown in the
dialog box (Figure 9-15).
Figure 9-15
Perspective
Calibration

3. Click View Linked CPT Files… to open the Calibration Image Analysis
dialog and the image display.
Figure 9-16
Perspective Calibration Screen
4. Select the camera and the calibration target image to view the
calibration point data.
5. Check Show Calibration Points… to view identified calibration grid
points. Check Show Grid Lines… to view grid lines in the selected
camera view,

Step E: Stereo AutoMapping
Stereo Automapping is a procedure for optimizing the perspective
calibration by correcting the misalignment error between the calibration
target and the light sheet. This section gives the step-by-step guide on how
to perform Stereo Automapping in Insight software.
What is Stereo Automapping?

Stereo Automapping is an optimization procedure used in Perspective
Calibration. This appendix introduces the theory of Stereo Automapping to
help you understand how it works.
Accurately placing the calibration target in the laser light sheet plane is the
largest error in most calibrations. The misalignment manifests itself as
mapping errors between world locations and the two camera images. The
mapping error, or disparity, can be measured by dewarping and cross
correlating the images. In dewarping, an image with a constant pixel
magnification is created by mapping from a world location into a camera
image and interpolating the camera image pixel intensity for each world
(dewarped) pixel location. The dewarped image is trapezoidal with the field
of view increasing with object distance. If the calibration was perfect and
the light sheet had zero thickness, the disparity between dewarped left
frame A and dewarped right frame A would be zero since both images
have captured the same particle field, but from two different perspectives.
The light sheet has some thickness which results in the disparity
correlation maps being elongated in the horizontal direction due to the
depth range within the light sheet.
100 μm 250 μm 500 μm 1,000 μm 1,500 μm

Figure 9-17
Correlation Peak Elongation with Increasing Light Sheet Thickness

The figures show how the correlation map peak width increases with
increasing light sheet thickness. The correlation maps were created from
very high concentration images. For typical particle image concentrations
the measured peak may be anywhere along the elongated peak as can
been seen in this disparity vector map.
Frame 001  29 May 2002  01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C2ACC8 | 01C
Aligned Target Vector Range

Figure 9-18
1 mm thick Lightsheet, 30 Vector Fields Disparity Vector
3 pixel
500
Field with Many
Single
450 Realization
Disparity
400
Vectors
350
Y pixel
300
250
200
150
100
50
0 100 200 300 400 500
X pixel
This disparity vector field shows all the vectors from a sequence with a
vector shown at each node from each realization. The range of vector
lengths matches the elongation of the peak. Ensemble Average Correlation
is the preferred averaging method for measuring the disparity with Insight
software. With ensemble processing the correlation maps are summed
together. When a sufficient number of realizations have been ensemble
processed, the correlation map peak will be filled in allowing accurate
location of the peak center position. If the light sheet is thin and particle
image concentration is high, you may be able to measure the disparity with
a single realization vector field.

The laser light sheet position in the target coordinate system can be
estimated from the disparity vector field. When the target is parallel to the
light sheet, but in front or behind the light sheet, the disparity vectors will be
uniform left or right. When the target has a rotation about the x-axis relative
to the light sheet, the disparity vectors will have a top to bottom gradient.
When the target has a rotation about the y-axis relative to the light sheet,
the disparity vectors will have a left to right gradient. When the target has a
combination of z-translation x-axis rotation and y-axis rotation, the disparity
map will have combination of uniform left or right with a top to bottom
gradient and a left to right gradient. Examples of these target positioning
errors are shown in the following disparity vector fields.
Frame 001  28 May 2002  D:ExperimentsTargetSimComponentSineVectorComponentSineAvgZ2target.vec
2 mm Z Target Postion
Figure 9-19
20 pixel Target at Z =
500
2 mm
450
400
350
Y pixel
300
250
200
150
100
50
0 100 200 300 400 500
X pixel
Frame 001  28 May 2002  D:ExperimentsTargetSimComponentSineVectorComponentSineAvgX2Target.vec
2 Degree Target X Tilt

Figure 9-20
Target X tilt =
7.5 pixel
500
2 deg
450
400
350
Y pixel
300
250
200
150
100
50
0 100 200 300 400 500
X pixel

Frame 001  28 May 2002  D:ExperimentsTargetSimComponentSineVectorComponentSineAvgY2Target.vec
2 degreee Target Y Tilt

Figure 9-21
7.5 pixel
Target Y tilt =
500 2 degrees
450
400
350
300
Y pixel
250
200
150
100
50
0
0 100 200 300 400 500
X pixel
Frame 001  28 May 2002  D:ExperimentsTargetSimComponentSineVectorComponentSineAvg222Target.vec
Target Position Z 2mm, X Tilt 2 deg, Y Tilt 2 deg

Figure 9-22
Target Z =
20 pixel
500
2 mm X tilt =
2 deg, Y tilt =
450 2 deg
400
350
Y pixel
300
250
200
150
100
50
0 100 200 300 400 500
X pixel
To correct the calibration target positioning errors, Insight software includes

the Stereo Automapping Processing Application. Stereo Automapping
dewarps the Left and Right A frame images, cross-correlates the two
dewarped A frames giving the ensemble disparity vector field, it analyzes
the disparity vector field to find the laser light sheet location with respect to
the calibration target coordinate system, and modifies the calibration file
with the corrected light sheet position.

The following gives an overview of the steps involved in the Stereo
Automapping procedure:
 Capture a sequence of particle flow images.
 Calibrate the system.
 Select StereoAutomapping processor.
 Select the calibration file.
 Select Dewarping Preprocessor.
 Select processor.
 Open raw images.
 Process to get light sheet position errors.
 Accept the correction to optimize the alignment.
 Open the raw images and repeat the processing to verify the
calibration accuracy.
(continued on next page)

To Verify and Optimize Calibration Target Location
1. Capture a sequence of particle flow images as described in Chapter 7
“Capturing Images.”
2. Calibrate the system as described earlier in this chapter.
3. From the Processing Tab select StereoAutomapping from the
Application drop list.
Figure 9-23
Application Screen

4. From the Preprocessing drop list select (Create New…).
Figure 9-24
Application / Preprocessing /
Postprocessing
5. Enter a Preprocessor name

or leave the default
“Preprocessor1” and click
OK.
Figure 9-25
Create New Preprocessor

6. Image dewarping is automatically selected as a preprocessor function.
You may also add other image preprocessing functions to the pipeline.
Click OK.
Figure 9-26
Processor Pipeline Editor

7. From the Processing drop list select (Create New…).
Figure 9-27
Processing
8. Enter the processor name

or leave the default name.
Click OK.
Figure 9-28
Create New Preprocessor

9. In the PIV Processor Setup dialog change Spot A and Spot B to 128.
Ensemble PIV is a fixed for StereoAutomapping. The other parameters
are the same as for PIV and StereoPIV processors and may be
changed for your experiment. Click OK.
Figure 9-29
PIV Processor Setup
10. Post-Processing is optional

but removing bad vectors is
usually a good idea. To set
up vector validation from
the Post-Processing drop
select (Create New…)
Enter a Post-Processor Figure 9-30
name or leave the default. Create New Postprocessor

11. Insert Local and/or Global Validation into the pipeline. With
StereoAutomapping it is helpful to remove bad vectors and having
holes in the vector field is okay. The light sheet position calculation
does a least-squares-fit of the disparity vectors to a plane. Smoothing
and filling holes might add errors to the least-squares-fit.
Figure 9-31

12. Select the calibration file from drop list. If you have not created a
calibration file, follow the instructions in “Perspective Calibration
Process” to create one.
13. Press the Start button to start processing.
Figure 9-32
Post-processing

Figure 9-33
Disparity Vector Field
14. Once the processing is done, the AutoMap dialog box will be
displayed.
15. The dialog box will display the light sheet location in target coordinates
if the correction is applied.
16. The dialog box will also display the change in light sheet location from
this Automapping iteration.
17. Disparity vector field statistics. The valid vectors denominator is the
grid size and does not discriminate by choice code.
18. Apply this Correction. Select OK to change the calibration. Cancel
leaves the calibration unchanged.

19. Click OK.
Figure 9-34
AutoMap Screen
20. To check how well the Automapping corrected the calibration errors,
click Start button to go through the process again.

Figure 9-35
Disparity Vector Field after Automapping
21. You may want to perform several iterations on the automapping,
particularly if the initial vector field is not very good. If the initial
disparity vector field has few vectors or indicates a large target error,
more iterations may be used.
22. Looking at the change in light sheet position fields in the Automap
dialog, shows when the AutoMapping has reached convergence.
23. A log of the AutoMapping changes is stored in the Settings folder in a
file name after the calibration file *_AutoMapLog.csv. This file can be
opened with Microsoft® Excel® or any ASCII file reader like Notepad.
To start a new log, delete the old csv file.
Figure 9-36
Viewing AutoMapping in Microsoft Excel Spreadsheet
Microsoft and Excel are registered trademarks of Microsoft Corporation.

Calibration for Measurements through a Window
In some experiments the flow is enclosed and the measurement must be
made through a window. If the access to the flow area is difficult, the
system may be calibrated on a bench top and then moved into position to
make measurements through the window. This procedure only works for
airflows.
The optimization procedure is the same as for standard light sheet position
correction in the previous section. A few tips for getting good results follow.
The bench top calibration could be performed with a similar window of the
same material, thickness, and position as the actual experiment. Using a
similar window will minimize the light sheet position errors that must be
corrected. The window should be large enough for both cameras to look
through. Uncorrectable calibration errors will result if two windows are used
and they are not in the same plane.
The calibration can also be performed with just air between the cameras
and the target. When the camera system is moved to the light sheet, there
will be an apparent z-axis translation due to the added glass between the
cameras and the light sheet. Make sure to perform the Automapping
procedure to get the proper calibration polynomials for this type of
calibration through windows.
If the correlation processing does not give good results on the dewarped
images, it may be because the disparity is too great. Larger disparities can
be processed by using the PIV Tab | Process Setup… | Processor
Settings… | Plugin Settings… | xSpotOffset plugin parameter. This
plugin parameter offsets the spots by input number of pixels. Entering the
approximate disparity will may give better results. The xSpotOffset may be
positive or negative.


CHAPTER 10
Processing and
Viewing PIV Images
Particle Image Velocimetry (PIV) measures the velocity of a fluid at many

locations in a plane. The fluid is seeded with small tracer particles that
move with the flow. A thin laser light sheet illuminates the particles in a
plane. A camera focused on the light sheet records the images of the
particles in the light sheet. By pulsing the light sheet twice, recording two
images of the particle field, the flow velocity can be estimated by
measuring the distance the particles have moved in the time between the
light sheet pulses.
Insight™ software allows you to process PIV images using both cross-
correlation technique and particle tracking technique. The cross-
correlation technique includes FFT correlation and ensemble correlation.
This chapter discusses how to:

 Overview of PIV Processing including the PIV Rules-of-Thumb.
 Process a PIV image using cross-correlation technique.
 Process a PIV image using particle tracking technique (super-
resolution particle velocimetry).
 View and analyze PIV results using TSI’s Tecplot® add-on and
MATLAB® toolboxes.
10-1
Rules-of-Thumb for PIV Analysis
This section gives some guidelines for optimizing your experiment for best
PIV results. To get the best results from a PIV experiment, the image
capture and the interrogation must match the flow. Typically, the
parameters that can be set to optimize the measurement are: the time
between laser pulses (dT); image shifting; interrogation spot size;
photograph magnification; light sheet dimensions; seed particle and seed
concentration. With these parameters in mind, the following rules were
developed.*
Rule 1 Interrogation spot size should be small enough so that one

vector describes the flow within that spot.
Rule 2 There should be more than ten particle image pairs per
interrogation spot.
Rule 3 Maximum in plane displacements should be less than ¼ of the
interrogation spot size.
Rule 4 Maximum out of plane displacement should be less than ¼ of
the light sheet thickness.
Rule 5 Minimum in-plane displacement should be two particle-image
diameters.
Rule 6 Exposure must be large enough to clearly show the particles.
These rules-of-thumb apply to the first pass of multi-pass processing. On

subsequent passes the previously measured vector field can be used to
optimize spot offset allowing smaller interrogation spots and larger
displacements relative to spot size. The methods for moving beyond the
rule of thumb limitations are listed as rule breakers.
*The PIV rules of thumb were developed by Keane and Adrian in Optimization of particle image velocimeters. Part 1: Double
pulsed systems Measurement Science and Technology 1 (1990). This paper included Rules 1 through 5, TSI added Rule
6.

Rule 1: Interrogation-Spot Size should be Small
Enough for One Vector to Describe Flow for that
Spot
This puts a limitation on the velocity gradient in an interrogation spot. If the
velocity gradient is too large then the displacement peak becomes larger in
area with a lower peak height, leading to the likelihood of measuring noise.
Figure 10-1 shows how measurement errors can occur because of too
large velocity gradients.
Velocity Gradient
Input Image Autocorrelation
Chosen Displacement
Particle Pairs
Not Overlapping
Figure 10-1
Velocity Gradient Error
Correlation produces the average velocity over the interrogation spot. The
larger the spot, the more the averaging that takes place. To see the fine
flow details, small spots are required. However, the minimum spot size is
determined by Rule 2.
Rule Breaker
Deformation processing allows larger velocity displacements within the
interrogation spot. The input spot is deformed according to the local
velocity gradient so that the particle displacement peaks from different
areas of the interrogation spot have the same displacement.
For boundary layer flows using an interrogation spot that is relatively long
and narrow reduced the velocity gradient in the interrogation spot. If the
spot area is held constant the number of particles in the interrogation spot
will also remain constant.
Processing and Viewing PIV Images 10-3

Rule 2: Should Have More than Ten Particle
Image Pairs Per Interrogation Spot
The percentage of correct velocity measurements as a function of number
of particle pairs has been established by Keane and Adrian. Their results
are shown in Figure 10-2. These results show that the percentage of
correct velocity measurements correlates strongly with the number of
particle image pairs in the spot. (This work also takes lost pairs, and Rules
3 and 4 into account.) As the concentration increases beyond one per spot,
the percentage of valid measurements increases. This is because many
pairs of particle images contribute to the displacement peak. The noise
peaks caused by the correlation of the images of different particles
decrease relative to the displacement peak. The loss of some pairs is
tolerable because many others pairs are still creating the
displacement peak.
Figure 10-2
Correlation of Correct Velocity Measurements with the Number of Particle Images
Particle image concentrations with an average of one per interrogation spot

produce the most erroneous measurements. This is because there is a
high probability of having two pairs, or one and a half pairs or two half pairs
in the interrogation spot. With two pairs of particles of equal size, the
correlation for the distance between the two particles is as strong as the
distance between the first and second image. With two pairs of particles of
equal size, the wrong measurement is made fifty-percent of the time. With
these low concentrations, a single particle that crosses the interrogation
spot boundary can cause wrong measurements to be made.

Figure 10-3 illustrates these errors for low particle-image concentration
images.
Interrogation Errors
Loss of Pairs Three Particle Images Two Particle Pairs
Measured Measured Measured

Disp. Disp. 1 of Disp. 1 of
3 Choices 2 Choices
Actual Disp.
Figure 10-3
Interrogation Errors
The percentage of correct velocity measurements increases with

decreasing particle image concentrations-from an average of one pair per
spot down to very low concentrations. This is because where a pair of
particles is found, it is unlikely that a second particle will also be in the spot
to cause confusion about the correct pairing. A velocity measurement is
only possible where a pair of particles is found so very low seeding
concentration pictures produce sparse vector fields. If the goal of the
experiment is to measure the flow on a high resolution grid, low seeding
concentrations must be avoided. This increasing correct vectors with
decreasing particle concentrations assumes that the processor is set to
reject areas with very low signals and that it can distinguish between the
gray scale variations in background light and the particle images.
Effective seeding can be increased by increasing the interrogation spot

size. With a larger interrogation spot there are more pairs per spot. When
you increase the spot make sure that you are not violating the velocity
gradient requirement of Rule 1. Decreasing the interrogation spot gives
more velocity vectors with the same interrogation spot overlap. For high-
resolution work and for investigating small flow structures, you should
select small interrogation spots.
Rule Breaker
With multi-pass processing the interrogation spots are offset by the image
displacement of the previous pass. By having a high percentage of
particles in the two spots pairs the required number of particles for a good
measurement is reduced.

Super-Resolution PIV uses the correlation processing results as an input to
assist in particle pairing. With super resolution good vectors are made on
individual particles.
Rule 3: Maximum In-Plane Displacements should

be Less than One-Fourth Interrogation Spot Size
This maximum displacement is a compromise between two parameters:
Large displacements for the best percent of reading accuracy and small
displacements to minimize lost pairs.
As explained before, the velocity is measured by finding the location of the

particle displacement peak. The location of the peak is measured with
centroid or curve fit to a fraction of a pixel. The accuracy of the PIV
measurement is a percent of full scale, or a number of pixels. The peak
location error is the same wherever it is located, near the center zero
velocity peak or near the edge. With large displacements the peak location
error is a smaller percent of reading than small displacement peaks. The
absolute maximum displacement that gives the correct displacement is half
the interrogation spot. If large displacements are used, the FFT aliases, the
peak location, and the measured peak give the wrong velocity.
Lost pairs refers to particles that are inside the interrogation spot for only
one laser pulse. In-plane motion causes particles to move across the spot
border. Fast moving particles are more likely to lose a pair than the slow
particles. Lost pairs creates a velocity bias towards slow velocity flow. The
lost pair estimation is given in Rule 4.
Lost Pairs, In Plane Motion
Lost Pair Area, Y Displacement

Displacement
dX
Vector
dY
First Images
Lost Pair
Y Spot in this Area
Area X
are Paired
Displacement
XSpot
%Paired = ( 1 - dX / XSpot) * ( 1 - dy / Y Spot) * 100%
Figure 10-4
Lost Pairs

The compromise between these two objectives is ¼ spot displacement, to
give good pairing and to use the accuracy of the system.
Rule Breaker
Setting the First Pass Offset parameter in the Grid Engine Plugin
Parameters allows particle pairing to be improved using a single offset
value for all interrogation spots. With the First Pass Offset the lost area is
reduced from the particle displacement distance to the range of
displacements.
With multipass processing the second and later processing passes the
spot offsets are optimized for each interrogation spot using information
from the previous pass displacement field allows the spot areas to be
aligned for minimum lost area.
Rule 4: Maximum Out-of-Plane Displacement

should be Less than One-Fourth of Light Sheet
Thickness
This rule is established to keep the lost pairs (due to out-of-plane motion)
to an acceptable level. In many experiments the light sheet is about 1 mm
thick and the interrogation spot is 1 mm  1 mm. With this 1 mm cube
measurement volume, the restrictions on out-of-plane displacement is no
greater than the in-plane displacement. With both in-plane and out-of-plane
displacements considered, the estimated percentage of particles with pairs
in the interrogation spot is:
Percentage Paired = (1–dX/Xspot) * (1–dY/YSpot) * (1–dZ/Thick) * 100%

where
Percentage Paired = Percent of particles with a pair in the
interrogation spot
dX = Particle image displacement in X direction
dY = Particle image displacement in Y direction
dZ = Particle image displacement is Z direction, out of
plane motion
XSpot = Interrogation spot size in X direction
YSpot = Interrogation spot size in Y direction
Thick = Light sheet thickness
When the maximum one-fourth interrogation spot and light sheet thickness
image displacement is used, an average of 42% of the particle images are
paired. The seeding concentration must be increased for the lost pairs to
get the same results. With one-fourth spot displacements in all three
directions then the required seeding concentration is ten pairs per spot
divided by the .42 paired, which is equal to 24 pairs per spot.
Rule Breaker
There is no processing improvement for out-of-plane displacements.

Rule 5: Minimum In-Plane Displacement should
be Two Particle Image Diameters
If the particle image displacement is less than two particle image
diameters, the image would be one elliptical particle image instead of two
particle images. The autocorrelation zero peak and displacement peaks
must be separated to accurately measure the centroid of the displacement
peak. If the peaks overlap, the centroid will include some of the zero peak
and bias the measurement towards zero displacement.
Figure 10-5 and Figure 10-6 show the autocorrelation at two points in a
turbulent flow. The displacement at the first spot is about two particle-
image diameters. In the autocorrelation, the displacement peak has
separated from the central zero peak. This point gives a good velocity
measurement.
Figure 10-5
Autocorrelation at Two Points in a Turbulent Flow

Figure 10-6 shows an interrogation spot with about one particle-image
diameter displacement. Here the displacement peak is connected to the
central zero peak producing a single oval central peak. When the
displacement peak blends into the zero peak, velocity measurement error
occurs. In this case the measured velocity is the distance between particles
and not the particle image displacement.
Figure 10-6
An Interrogation Spot with One Particle-Image Diameter Displacement
Rule Breaker
This rule is for autocorrelation processing only. Two-frame cross
correlation processing removed this limitation. There is no problem with
having the first and second particle images at the same location with two
image frames.

Rule 6: Exposure should be Large Enough to
Show Particles
There are many factors that determine the exposure. These are:
 Laser energy
 Light sheet dimensions
 Camera Focal length number
 Photograph magnification
 Film Speed and resolution
 Particle size, and
 Particle material
Regarding particle scattering, particle size and material determine the

scattered light power in the image of each particle. The particle scattering
is in the Mie regime for most PIV applications. Mie scattering is complex
and the scattered energy varies greatly with scattering direction, particle
size, and particle index of refection. The complexity of Mie scattering
makes it difficult to make general statements about how much exposure is
required for a given experiment. In most applications through trial and
error, you can find the correct exposure.
It is difficult to quantify the parameters that affect exposure and come up

with a clear recommendation. Generally, a 5-Watt Argon Ion laser can be
used for water flows that have low velocities. Use a Nd:YAG laser for
higher velocity flows, especially air flows with micron -sized seed particles.
Rule Breaker
Image Preprocessing with background subtraction and filtering can help
make the most of the images you have. The ability to enhance a poor
image is limited. For best results capture good images.
If the images are dim the image quality can generally be improved by using
more laser power, a smaller laser light sheet, a larger camera aperture, or
a more sensitive camera.
If image quality is poor because of background scattered laser light using

fluorescent particles and a band-pass filter on the camera may help.

PIV Cross-Correlation Processing
Cross-correlation processing of PIV images can be broken down into these
steps:
 Image preprocessing,
 Generating grids
 Masking spots
 Performing correlation
 Locating peaks and
 Performing vector validation
The flow chart shows the processing steps and how a vector field is
generated from the input images and the process data. There are several
algorithms that could be used for the processing steps: grid generation,
spotmasking, correlation and peak location. The best choices depend on
the flow, velocity gradients, seeding, and allowable processing time. Insight
software is designed to enable you to match the algorithms to your flow.
Insight software comes with several algorithms to choose from for each of
the processing steps. These algorithms produce good results for the
majority of experiments. However, to facilitate measurements in flows with
special challenges and research, user -created algorithms can be plugged
into one of these processing points. See Chapter 18, “Developing and
Using Plugin Dynamic Link Libraries (PIV Only).”

Input Figure 10-7
Images PIV Processing
Flowchart
Image
Conditioning
Image
Yes Processor
Conditioning?
(Background
Subtraction)
No
Grid Engine
Preliminary
Generate Spot
Vector
List & Initialize
Field
Vector Field
Spot Mask
Engine
Condition Spots
Correlation
Engine
Compute
Correlation Map
Yes
Peak Engine
Measure
Displacement
Store Vector
More Spots?
No
Another
Pass Validate Yes
Pass?
No
Final Validate
Store Vector
File
Send Tecplot
Message

Setting up PIV Processor
Before processing PIV images, you need to set up the processing
parameters.
1. Select an image from the Exp. Tree, for which you want to setup the
processing parameters. Drag and drop it onto the display panel. See
Chapter 6, “Viewing, Enhancing, and Displaying Image Files” for
detailed information on how to display and enhance images for
processing.
2. Select the PIV or Stereo PIV Application in the processing tab. In the
Processing selection box, select Create New to create a new PIV
processor (or select an existing processor to edit it).
3. Click Setup button, the PIV Processor Setup dialog opens
(Figure 10-8).
Figure 10-8
PIV Processor Setup
4. Select the Classic PIV or Ensemble PIV algorithm.
5. To use Deformation Processing Algorithm, simply check Use Image
Deformation checkbox.
6. If Normalize Image Intensity checkbox is checked, each frame is
normalized using the minimum and maximum intensity before the
correlation analysis.
7. Select the plugin engines for PIV processing. Click Plugin Settings to
edit plugin engine settings. Please see PIV Plugin Engines for detailed
description of plugin parameters and plugin data.

8. If you have selected a grid engine that does recursive processing, you
need to setup a pass validation to validate the velocity data between
the passes. Click Pass Validation Settings to setup a vector
validation pipeline. Please see PIV Vector Validation for detailed
description of different vector validation methods.
9. In the Spot Size boxes, specify the dimensions of the interrogation
spots in pixels. For recursive processing, you also need to specify the
final spot size. Check Square checkbox to use square spots.
10. In the Maximum Displacement box, specify the maximum particle
image displacement, as a fraction of the current spot size that will be
looked for in processing. The Peak Engines use this value to limit the
correlation map search area to the maximum displacement pixels from
the zero. Some Correlation Engines use this value to limit the size of
the correlation map to increase processing speed and decrease
memory usage.
11. Click Advanced button will open PIV Advanced Settings dialog
(Figure 10-9).
Figure 10-9
PIV Advanced Settings
12. Enable Double Correlation in the Advanced Settings dialog. Please

see Double Correlation Analysis for more information about these
algorithms.

13. Select the number of secondary correlation peaks exported to the
vector files. The secondary peaks can be used in Vector Validation to
replace bad vectors of which the highest peak fails the validation.
14. Double Correlation is a special vector validation method that requires
correlation map data. It compares the original vector with the vector
obtained from the multiplication of the correlation maps of the current
vector and its neighboring vectors. If the difference between the two
vectors is greater than the given tolerance, the current vector is
labeled as a bad vector.
15. Turn off Multi-thread Processing here. Turning this off is sometimes
useful for troubleshooting problems with a computer’s multiple
processor setup, but will slow down processing.
16. Default Delta T: Displays the default delta T setting that was
embedded during image capture. For images captured with previous
version of Insight software, specify a value in this box for processing
your PIV vectors in Velocity mode.
17. Override TIFF Tag Delta T: When images not captured using Insight
software are processed, using PIV, PTV, or GSV, they may not contain
the delta time information of the laser pulses. When this is selected,
the entered delta time is used to process these images.
18. PIV Uncertainty export options are controlled here. There are two
levels of uncertainty: Standard Uncertainty and Expanded
Uncertainty. See PIV Uncertainty for more detailed information.
PIV Plugin Engines

Insight software implements PIV processing algorithm in the following four
plugin engines:
Grid Engines For grid generation.

Spot Mask Engines For masking spots.
Correlation Engines For cross-correlation.
Peak Engines For finding correlation peaks.
Grid Engines
The GridEngine breaks the input images up into smaller spots for
processing and initializes the vector field. For each grid point, the process
manager copies the pixels from the input images into the spots and passes
the spots to the SpotMaskEngine.
The GridEngine can process the vector field in single as well as in multiple
passes. In a typical multiple-pass processing, on the first pass, the spots
are set to follow the ¼ displacement rule of thumb with no spot offsets.
After the first processing pass, the vector field from the previous pass can
be used to optimize the spot grid for the next processing pass. By offsetting

the spots by the particle image displacement distance, lost pairs due to in-
plane motion are eliminated increasing the correlation signal-to-noise ratio.
Offsetting the spots by the particle image displacement distance also
allows for smaller spots since the ¼ displacement rule no longer applies.
Using recursive grid algorithms allow for more accurate and higher density
flow measurements.
Plugin Description
Nyquist Grid Use this plugin for the fastest results.
It is the classic PIV grid. It sets vectors with the x
spacing equal to half Spot A width and the y spacing
equal to half Spot A height. No spot offsets are used
and the processing uses only a single pass. This gives
a vector grid with 50% spot overlap fitting the Nyquist
sampling criteria.
Recursive Nyquist Use this plugin for increased accuracy or higher spatial
Grid resolution.
It processes the images in two or more passes. The first
processing pass computes the vector field at the
starting spot sizes with Nyquist, 50% overlap grid
spacing. The vector field is then edited using the Pass
Validation. The results of the first processing pass are
used to optimize the spot offsets for the second
processing pass. The offset between the spots is equal
to the integer pixel displacement measured in the first
processing pass so that subsequent processing passes
would have a peak location within a half pixel of the
correlation center. If the final spot sizes are smaller than
the starting spot sizes, the spot size are reduced by a
factor of two and the number of vector rows and
columns increased by a factor of two, giving four times
the number of vectors with each pass. If the starting
and final spot sizes are the same, the second pass
uses the optimized window offsets to recompute the
vector field with the same number of vectors.
Rectangular Grid Use this plugin for full control of the recursive grid
settings.
It provides more user grid parameter control than the
Nyquist or Recursive Nyquist Grids. With the default
settings, the grid is the same as Recursive Nyquist
Grid. If the Recursive setting is changed to false then it
is the same Nyquist Grid. This grid has settings to
change the spot spacing to values other than 50%
overlap and to change the spot size reduction factor
between processing passes.

Spot Mask Engines
The SpotMaskEngine conditions the spots for processing and passes the
spots to the CorrelationEngine. It is used to modify or condition the image
spots before processing. This can increase the signal-to-noise ratio of the
correlation map.
Plugin Description
Gaussian Mask With Gaussian Mask, each pixel in spot A is multiplied
by a Gaussian weighing function so that the spot is
bright in the center and dark around the edges. This
weighting gives more value to the center pixels and less
to the edge pixels.
ZeroPad Mask The ZeroPad Mask computes the average pixel
intensity of each spot and subtracts it from each pixel. If
the size of input spots is not a power of 2 or if the spot B
is larger than spot A, the spots are padded with zero
intensity pixels to a square power of 2 (8  8, 16 16,
32  32, 64  64 or 128  128). For example if Spot A is
15  10 pixels and Spot 2 is 32  32 pixels, Spot A will
be padded to 32  32 pixels and Spot 2 will not be
padded. The advantage of zero padding is increased
signal-to-noise ratio. The disadvantage is increased
processing times.
No Mask When No Mask is selected, the original spot image is
sent to the correlation engine for cross-correlation
analysis.
Correlation Engines
The CorrelationEngine computes the correlation function and returns it as
a correlation map. It computes the correlation function of the masked Spot
A and Spot B returning the result as a correlation map. The correlation
function is an algorithm that sums the particle image matches at all pixel
displacements within the displacement range. The highest correlation map
pixel is assumed to be the particle image displacement peak caused by the
contributions of many particle pairs. Other peaks are assumed to be noise
peaks caused by the random pairings of images of different particles.
Correlation algorithms may generate peaks with different sizes, shapes, or

slopes. The measurement accuracy may be improved by matching the
peak algorithm to the correlation algorithm. The CorrelationEngines may
set a default PeakEngine that is automatically selected each time the
CorrelationEngine is selected.
Plugin Description
FFT Correlation It computes the correlation using Fast Fourier
Transform (FFT). The spots must be square powers of 2
and the spot A must be the same size as the spot B.
The ZeroPadMask must be used when the two spots
have different sizes.

Plugin Description
Direct Correlation It computes the correlation by first computing and
subtracting the spot mean intensity and then summing
the product of pixel intensities at each pixel offset. The
correlation map is then adjusted by an autocorrelation
factor to increase the displacement measurement
accuracy. It is the slowest correlation engine.
The source code of Direct Correlation Engine can be
found at <Insight>\PluginSDK\ExampleCode\
DirectCorrelator. Users who are interested in writing
their own PIV plugin engines can use this code to learn
how to develop plugin engines.
Note: CorrelationEngine has the Default PeakEngine
PluginParameter to define the PeakEngine that
works the best with the current correlation
engine.
Peak Engines
The PeakEngine finds the peak location in the correlation map.
Plugin Description
Gaussian Peak It locates the correlation peak with sub-pixel accuracy
by fitting a Gaussian curve to the highest pixel and its
four nearest neighbors. Two 3-point fits are done: one
in the x direction with the peak pixel and the pixels to
the left and right of the peak; and one in the y direction
with the peak pixel and the pixels above and below the
peak.
The Gaussian peak equation is given by:
dx  x 
log(l )  log(r )  x0
2  log(l )  log(r )  2 log(c )
where l, r, and c are intensity value for left, right and
peak pixels in the correlation map, x is the integer shift
and x0 is the zero shift location.
Gaussian peak is the recommend peak engine for use
with FFT Correlation Engine and Direct Correlation
Engine.
Bilinear Peak It locates the correlation peak with sub-pixel accuracy
by fitting a set of linear functions to the highest pixel
and its 4 nearest neighbors. Two 3-point fits are done:
one in the x direction with the peak pixel and the pixels
to the left and right of the peak; and one in the y
direction with the peak pixel and the pixels above and
below the peak.
The bilinear peak equation is given by:
(r  l )
dx  x   x0
2  max( c  l , c  r )
where l, r, and c are intensity value for left, right and
peak pixels in the correlation map, x is the integer shift
and x0 is the zero shift location.

Edit Plugin Parameter and Plugin Data
Each plugin engine can have a number of parameters that can be used to
configure the algorithm. These parameter values are used as input during
processing. Each plugin can also have data values that are measurements
made during processing for each grid point, and the data can be exported
to the vector files.
Plugin parameter and plugin data can be edited in the PIV Plugin Engine
dialog, which is opened by clicking the Plugin Settings button is the PIV
processor dialog.
Note: It is recommended that you use the default parameter and data
values and settings until you have processed a few images and
have assessed your experiment needs properly.
Figure 10-10
PIV Plugin Engine Settings
Note: The Plugin Engine Settings dialog has the description of every
plugin parameter and plugin data. The same information is also
given below for your reference.

Grid Engine Parameters
Grid Engine: Nyquist Grid
Parameters Value and Description
NyquistGrid: The first pass-offset parameters allow you to
XSpotOffset, YSpotOffset enter a spot offset for the initial processing pass.
In the second and later passes, the spot offset is
set using the previous pass vector field. First
pass offset is most useful in flows with a large
mean velocity and small variations from the
mean.
Grid Engine: RecursiveNyquist Grid

RecursiveNyquistGrid: True: The spot offset algorithm moves both Spot
Central Difference Offset A and Spot B by one-half the particle image
displacement measured in the previous
processing pass, relative to the vector (x, y)
location.
False: Spot A is centered on the vector (x, y)
location, Spot B is offset by the particle image
displacement distance.
RecursiveNyquistGrid: The first pass-offset parameters allow you to enter
xOffsetPass1 (Default = 0) a spot offset for the initial processing pass. In the
yOffsetPass1 (Default = 0) second and later passes the spot offset is set
using the previous pass vector field. First pass
offset is most useful in flows with a large mean
velocity and small variations from the mean.
Grid Engine: Rectangular Grid

RectangularGrid: True: (Default) Enables multi-pass processing.
Recursive False: Sets single-pass processing.
RectangularGrid: Set the distance between grid points based on the

xSpacing(spotWidth) and spot width and height.
ySpacing(spotHeight): Note: Using the default value of 0.5, gives the
same 50% spot overlap as Nyquist. Using values
smaller than 0.5 gives oversampling. Values larger
than 0.5 gives undersampling.
RectangularGrid: Select the amount by which the spot size is
SpotReduction: (Default=2.0) reduced between processing passes. With the
default 2.0 the spot size is ½ the width and ½ the
height of the previous processing pass increasing
the number of vectors 4 times. If the
SpotReduction were increased to 4 the spotWidth
and height would be ¼ the previous spot size and
the number of vectors will increase by 16x.

Grid Engine: Rectangular Grid
RectangularGrid: CenterOffset: True: Both Spot A and Spot B are displaced by ½
(Default: True) the offset distance. The vector is located halfway
between the two spots.
False: Spot A is centered on the vector location
and only Spot B is offset.
RectangularGrid: The first pass-offset parameters allow you to enter

xOffsetPass1 (Default = 0) a spot offset for the initial processing pass. In the
yOffsetPass1 (Default = 0) second and later passes the spot offset is set using
the previous pass vector field. First pass offset is
most useful in flows with a large mean velocity and
small variations from the mean.
Grid Engine: Deformation Grid

PluginData Value and Description
Deformation Grid: The distance, in pixels, between vectors.
XGridSpacing
YGridSpacing
Deformation Grid: The interger interrogation spot offset between
xSpotOffsetPixel Spot A and Spot B. This controls the input pixels
ySpotOffsetPixel for the spots before deformation.
Deformation Grid: Instructs Deformation Spot Mask to interpolate
Interpolate and deform the spot or leave the input spot as is.
Deformation Grid: The velocity vector and the four corner vectors
dXCenterVector from the previous processing pass.
dYCenterVector
dXLeftTopVector
dYLeftTopVector
dXLeftBottomVector
dYLeftBottomVector
dXRightTopVector
dYRightTopVector
dXRightBottomVector
dYRightBottomVector
Deformation Grid: Spot offset used for first processing pass.
XOffSetPass 0.0
YOffSetPass 0.0
Deformation Grid: True: Use Center Spot Offset
CenterOffset False: Use forward Offset.
Primary and Secondary Number of processing passes at starting and final
Iteration spot size.

Spot Mask Engine Parameters
Spot Mask: Gaussian Mask
Gaussian Mask: Minimum Use to filter out image spots that do not have any
Average Spot Intensity particle images. Set so that areas without particle
images are below the threshold and spots with
particle images are above the threshold. Use
Point Processing and the Process Monitor to help
you find the correct threshold value for your
images.
Subtract Mean Subtracts the spot mean value from each pixel
before multiplying by the Gaussian weighting
factor. Default = true.
Gaussian Mask: Gaussian Gaussian Mask has one PluginParameter—
Radius Gaussian diameter in spot sizes. The default
Gaussian diameter, 1.0, reduces the intensity at
edges of the horizontal and vertical axes to 13.5 %
of the center pixel.
Spot Mask: ZeroPadMask

ZeroPad Mask:Minimum Filters out image spots that do not have any
Average Spot Intensity particle images. The average spot intensity
should be set so that areas without particle
particle images are above the threshold. Using
Point Processing and the Process Monitor will
help you find the correct threshold value for your
images.
Spot Mask: No Mask

No Mask:Minimum Average Filters out image spots that do not have any
Spot Intensity particle images. The average spot intensity
should be set so that areas without particle
particle images are above the threshold. Using
Point Processing and the Process Monitor will
help you find the correct threshold value for your
images.

Correlation Engine Parameters
Correlation Engine: FFT Correlator
Plugin Parameter Value and Description
IsSpotZeroPadded Was the spot zero padded? The correlation map
for non-zero padded spots is adjusted to account
for the number of pixels that can be correlated to
create the displacement peak.
Default =false.
Default Peak Engine: The GaussianPeak is the default PeakEngine for

Gaussian Peak FFT Correlation. When the FFTCorrelation is
selected, the PeakEngine is set to the default
PeakEngine setting.
Correlation Engine: Direct Correlator

Plugin Parameter Value and Description
Default Peak Engine: The GaussianPeak is the default PeakEngine for
Gaussian Peak Direct Correlation. When the Direct Correlation is
selected, the PeakEngine is set to the default
PeakEngine setting.
Peak Engine Parameters

Correlation Engine: Gaussian Peak AND Bilinear Peak
Peak-to-Noise Ratio The base peak class used to derive
PeakEngines from finds the largest peaks in the
correlation map. The highest peak is used to
locate the particle image displacement. The
second highest peak is a noise peak caused by
the random pairing of images from different
particles. The ratio of the displacement peak
height to the noise peak height shows how much
the displacement peak stands out above the
noise. Low Peak/Noise peak values indicate that
there is less confidence that the selected peak is
the correct peak.
Note: When the SNR Filter Enable is checked in the PlugIn Data
Configuration Editor, the measured must pass the threshold test to
be included in the vector field. If the value does not pass the
threshold test the vector is marked with choice code 0 (SNR_FAIL).
The best ways to set the SNR threshold is to watch the Process
Monitor with Point or Step processing or view the exported data in
Tecplot® to see what the values are.

Double Correlation Analysis
Double Correlation
Double Correlation is a vector validation method built into the PIV
processor, like the SNR validation in the peak engine. It compares the
original vector with the vector obtained from the multiplication of the
correlation maps of the current vector and its neighboring vectors. If the
difference between the two vectors is greater than the user-defined
tolerance, then the original vector is set as a bad vector.
(continued on next page)

PIV Process Monitor
When you select Process Monitor on the processing tab, the PIV Process
Monitor is displayed as the top-most window during processing to allow
you to see what is happening during the processing. The Process Monitor
dialog can be very helpful when trying to understand how the experiment is
working, or why it is not working. With the Step Processing function, it also
allows you to pause the processing after each spot is processed.
Figure 10-11
PIV Processing Monitor
The Process Monitor shows the following data of PIV processing:

 The first two image spots are the Spot A and Spot B images as taken
from the input images.
 The second two image spots are the Spot A and Spot B images after
processing showing changes that may have been made by the
SpotMask and CorrelationEngines. For example, if ZeroPadSpotMask
is selected, Spot A will have the edges blacked out.

 The correlation map can be viewed as a 2-D image or a 3-D map.
When a 3-D map is shown, right-click on the map and select Maximize
to view the 3-D map on the full screen. Using the scroll bar you are
able to change the view position to see the 3-D map from different
angles. The right-click menu also allows you to export the 3-D map as
an image file.
 The plugin engine information and the vector data are also shown in
the monitor.
Here are a few tips on how to use the processor monitor in the step
processing mode to help optimize the PIV processor settings:
 Looking at the number of particles in the spots is helpful in optimizing
the spot sizes. Looking at the Masked Spots helps in seeing if the
SpotMask and CorrelationEngines are improving the spot image
quality.
 In the Correlation map look to see if the particle image displacement
peak stands out from the background noise. Does the peak move
smoothly when an area is processed? Is the peak inside the center ½
of the correlation map or is near the edge? If the peak is too close to
the edge you will need to start with a larger spot size or use a first pass
offset.
 In the PluginData check to see which parameters have failed. Seeing
the values for good vectors and bad vectors will help in choosing an
effective threshold value.
Particle Tracking Analysis (Super-

Resolution Particle Velocimetry)
Insight software allows you to process PIV images using Particle Tracking
technique. Particle Tracking finds the velocity of individual particles, versus
a group of particles as used by the cross-correlation technique.
Insight software uses an advanced particle tracking algorithm based on the

ICP (Iterative Closest Point) algorithm that is widely used in Computer
Vision. Insight software integrates PIV processor with the particle tracking
algorithm seamlessly to provide you the Super-resolution Particle
Velocimetry processor for super-resolution analysis of PIV images.
Super-resolution Particle Velocimetry processing can be broken down

into these steps:
1. Performing image preprocessing (optional).
2. Particle Identification: Find particles in the images using particle
identification algorithm.
3. Local Flow Calculation: Estimate local flow field using PIV processor.

4. Particle Tracking: Find particle velocities using particle tracking
algorithm.
5. Velocity Interpolation: Calculate grid velocities using velocity
interpolation algorithm.
6. Performing validation of grid velocities (optional).
The flow chart shows the processing steps of Super-resolution Particle

Velocimetry processor, and how to obtain two particle files and a vector
file from the input image pair.
Input
image pair
Image Needs
Yes
Conditioning Preprocessing?
Processor
No PTV Processor
Particle Save
Identification PAR files
PIV Yes Local

Processor Flow Estimation?
No
Particle Save
Tracking PAR files
No Get
Grid velocities?
Yes
Velocity Save
Interpolation VEC files
(optional)
Final Validation
Send Tecplot
Message
Figure 10-12
Super-resolution Particle Velocimetry Processor Flowchart

Image Preprocessing
Preprocessing particle images can improve the performance of particle
identification. For example, when particles are less than three pixels in the
images, particle identification often fails to find the particles. This problem
can be solved by applying Gaussian blurring to the raw images. Image
preprocessing can also be used to eliminate the background and remove
non-uniform illumination. See Chapter 8, “Processing Images” on how to
setup and apply an Image Conditioning Processor.
Setting up a Processor
1. Select an image from the Experiment Tree, for which you want to setup
the processing parameters. Drag and drop it onto the display panel.
2. Select Super-resolution Particle Velocimetry application in the
processing tab.
3. In the Processing drop-down list, select (Create New…) to create a
new processor, or select an existing processor to edit.
4. Click Setup button to open the processor setup dialog (Figure 10-13).
Figure 10-13
Super-resolution Particle Velocimetry Processor Setup
Particle Identification
Particle Minimum Size The minimum size of particles to be found in the
(pixels) image. Objects of smaller size are rejected.
Particle Peak Intensity The minimum peak intensity of a particle. Unless
Threshold adaptive thresholding is used, the threshold entered
should be a grayscale value between 1 and 255 for 8-
bit image, between 1 and 4095 for 12-bit image, and
so on.

To find the optimal threshold, you may use Threshold Assistant to see how
different threshold values affect particle identification.
Figure 10-14
Threshold Assistant
Local Flow Estimation

1. For super-resolution analysis, you need to setup a PIV processor for
calculating the local flow field to guide particle tracking. It is also
recommended to setup a vector validation to validate the PIV result.
Please see “Setting up PIV Processor” and “PIV Vector Validation” for
help on how to set up PIV and vector validation.
2. To skip local flow estimation step, check the Do Not Use Local Flow
Estimation check box.
3. If there is existing PIV vector file that you want to use, check Use
Existing PIV Vector Files to load vectors from the most recent PIV
vector files (.VEC) of the current image. The processor will skip the
local flow estimation step.
IMPORTANT
The history logging needs to be enabled in order to use existing PIV vector
fields. Otherwise, vector files generated by the PIV processor for the same
image may be overwritten by the Super-resolution Particle Velocimetry
processor.

Particle Tracking
Particle Group Size Particle tracking algorithm operates on a group of
particles. A minimum of 16 particles per group is
required.
Pass Validation Each particle in a group is shifted according to local
Tolerance flow field, and the distance to its nearest neighbor
(called deviation) in the second frame is calculated. A
particle pair is matched if its deviation is less than the
tolerance times the median deviation of the group.
Final Validation The velocity of each matched particle is checked by a
Tolerance median filter using the velocity of its matched
neighbors. Match validation parameter defines the
median filter tolerance.
Velocity Interpolation
Velocity Interpolation calculates the velocity on a regular grid so that
derivate quantities of velocity such as vorticity and shear stresses can be
calculated. Starting with randomly distributed particle velocities, Delaunay
Triangulation is first used to find neighbor particles of each grid points, and
then the bi-cubic interpolation is used to calculate the velocity at each grid
point using the velocity of its neighbor particles. The output is a vector file
(.VEC) which is identical to a PIV vector file except for a PTV tag in the
Tecplot® header.
Create vec file by Enable/disable velocity interpolation. No grid vector file
vector interpolation is generated if velocity interpolation is skipped.
Velocity Grid Interval The interval between the grid points, which defines the
(pixels) spatial resolution of the grid vector data.
Max Distance to Each grid point must have at least one neighbor
Nearest Particle particle within this maximum distance. Otherwise no
(pixels) velocity is calculated at this grid point.
PIV Vector Validation

The processing of PIV images can produce over 95% correct velocity
measurements if the guidelines for PIV image capture and processing
developed by Keane and Adrian (refer to the PIV Reference manual) are
followed. Lost pairs due to in-plane and out-of-plane motion, or low seeding
density causing low correlation signal strength lead to spurious vectors.
Spurious vectors happen when the highest correlation peak is not due to
pairs of particles moving with the flow, the velocity peak, but when, instead,
a random pairing of particle images produces a signal highest correlation
peak. Vector field validation functions are designed to filter out the spurious
vectors and fill the removed vectors through interpolation.

Insight software performs vector validation as a post-processing step.
Vector validation can also be launched during recursive PIV processing
between the passes. The steps of setting up final validation and pass
validation are identical, as described below.
Setting up Vector Validation

1. For final validation, go to the processing tab and select (Create
New…) or a saved setting file in post-processing drop-down list. Vector
validation is a post-processing step in PIV, Stereo PIV, Stereo
Automapping and Super-resolution Particle Velocimetry applications.
Click Setup button to open the vector validation pipeline editor.
2. For pass validation, click Pass Validation Settings button in PIV
Processor Setup dialog.
3. The Vector Validation Pipeline editor opens (Figure 10-15).
Figure 10-15

4. In the processor modules dialog box, select vector validation processor
of your choice and add them to the pipeline by double-clicking, drag-n-
drop, or using the Add button.
Figure 10-16
Processor Pipeline Editor Select Vector Validation Processor
5. Setup individual processors in the pipeline by double-clicking the icons
or using the Setup button in the toolbar.
6. Click OK with you are done.
7. For final validation, the currently selected setting file will be used when
Start button is clicked.
8. For pass validation, the settings are saved as part of PIV processor
setting file.

Vector Local Validation
Vector Local Validation uses the vectors in the neighborhood of each
vector to calculate a reference vector for validation. When the difference
between the current vector and the reference vector is greater than the
user-defined tolerance, the current vector is invalid and the vector choice
code (CHC) is set to -1 (Failed Validation).
The setup dialog of Vector Local Validation opens (Figure 10-17).
Figure 10-17
Vector Local Validation Setup
There are three local validation methods available, based on how the
reference vector is calculated:
Validation Method Description

Local Median This is the most widely used local validation method. The
velocity components of the reference vector are the median
value of all vectors in the neighborhood. The user-defined
tolerance, in the pixel unit, sets the maximum allowed
difference between the current vector and its reference
vector.
Local Mean Similar to Local Median, except that the reference vector is
the mean of all vectors in the neighborhood.
Universal Median The reference vector is the same as that in Local Median
method. Instead of directly comparing the velocity
magnitude of the vectors, this method normalizes the
difference between the current vector and the reference
vector by the median of the differences between all the
neighborhood vectors and the reference vectors. The
normalized difference, which is dimensionless, is then
compared with the user-defined tolerance. The advantage
of Universal Median method, as its name implies, is that
the tolerance can be applied to the whole velocity field
regardless of the local flow characteristics.
Vector Local Vector also allows you to replace the invalid vectors by either
the local median vector or a valid secondary peak when the secondary
peaks are exported in PIV (in PIV Advanced Settings Dialog). The replaced
vectors have choice code of 4 (interpolated).

Note: When both checkboxes are checked, Replacement by Secondary
Peak has the priority over Replacement by Local Median.
Vector Global Validation

Vector Global Validation is a range filter that detects vectors outside a
user-defined valid velocity range. It is a global filter because the velocity
range is applied to the whole velocity field.
The setup dialog of Vector Global Validation opens (Figure 10-18).
Figure 10-18
Vector Global Validation Setup
There are two global validation methods available, based on how the range
is defined:
Absolute Range The range of valid velocity is defined by the
velocity magnitude.
Standard Deviation Range The range of valid velocity is defined by the
multiply of standard deviation and is centered
at the mean velocity.
If a vector field is available in the current image window, the velocity

statistics of the current field is also show in the dialog.

Vector Conditioning
Vector Conditioning is usually used after vector validation to fill holes in the
vector field and/or to smooth the vector field. The holes can be due to
failed SNR in PIV processing or failed validation in previous validation step.
The setup dialog of Vector Conditioning opens (Figure 10-19).
Figure 10-19
Vector Field Conditioning Setup
Filling Holes
The holes in the vector field will be filled with the mean or median vector of
the neighborhood vectors. The filled vectors have the choice code of 4
(interpolated).
Note: A minimum of three good neighbor vectors are required to fill
a hole.
Validation Method Description

Recursive Filling When it is checked, the filling procedure sorts the holes by
the number of valid neighbors found initially. It first fills the
holes with the most valid neighbors since they have the
best chance to be filled; it then fills the holes with the
second most valid neighbors, in which the holes filled in the
previous pass are also treated as valid neighbors; the
same process is repeated until all holes are processed.
Note: Without using the recursive filling, the filling always
starts from the top-left corner of the vector field, and moves
toward the lower-left corner row by row. This may results in
unwanted artificial pattern in the resultant vector field if
there are many holes in the input.
Use measured When it is checked, only vectors originally obtained by the
vectors only PIV processor will be used as good neighbor vectors.
Other good vectors, such as those replaced in local
validation or in previous conditioning steps, are excluded.

Smoothing
Smoothing replaces every vector by its Gaussian-weighted mean of the
neighbor vectors. It is equivalent to a low-pass filter. After smoothing all
vectors’ choice code is set to 5 (smoothed).
PIV Uncertainty
Conceptually, the Uncertainty of Measurement can be described as a level
of doubt in a measurement. Practically, the uncertainty is manifested as a
numerical bound in which the true value lies, with some level of confidence.
This is graphically represented in the figure, where the true value on the
number line is within the bound of uncertainty.
Figure 10-20
Uncertainty on a Number Line
Method
Insight software uses the Peak Ratio (PR) uncertainty method, detailed in
Charonko & Vlachos1. This method uses the Peak to noise Peak Ratio
(PPR) to determine the uncertainty. The PR method incorporates many
possible sources of error (e.g., particle seeding density, pixel displacement,
image pre-processing, etc.), taking a global approach to calculating
uncertainty. In general, the uncertainty is inversely proportional to the PPR.
The PR method has already been calibrated for use in Insight software and
requires no calibration by the user.
The PR method assumes that errors in pixel displacement follow a

Gaussian distribution. Therefore, the uncertainty must be expressed with a
corresponding confidence level. In Insight software, uncertainties can be
exported at the Standard and/or Expanded level(s):
 The Standard Uncertainty implies that 68% of vectors will have their
true value lie within the uncertainty bound.
 The Expanded Uncertainty implies that 95% of vectors will have their
true value lie within the uncertainty bound.
1 Charonko J. & Vlachos P., 2013, Estimation of uncertainty bounds for Individual particle image velocimetry measurements from cross-
correlation peak ratio, Measurement Science Technology, 24.

Component-wise Uncertainties
Insight software calculates two-dimensional velocity vectors when using
the PIV application, and therefore, each component of velocity will have its
own uncertainty, and , at some confidence level. Importantly, the
uncertainty exported by Insight software, is a scalar value—the magnitude
of the uncertainty. Graphically, this can be visualized as:
Figure 10-21
A Graphical
Expression of the
Magnitude of
Uncertainty (Red
Circular Region)
where ⃗ is a velocity vector and is the magnitude of uncertainty at some

confidence level. The red circle represents spatial bound where the head of
the true velocity vector could possibly exist.
Using the PR method, it is assumed that . Therefore, if you would

like to calculate the uncertainty for a particular velocity component, simply
divide the uncertainty magnitude by √ , since ( ).
For example, to calculate the Expanded (95%) Uncertainty for either
component, use the following equation:
Insight software exports the To obtain an uncertainty for a

uncertainty magnitude at a particular velocity component,
given confidence level. simply divide by the √
The PR method determines uncertainties for two-dimensional PIV. And

thus, the reconstructed three-dimensional velocities obtained using Stereo
PIV will not receive a corresponding uncertainty. However, because two-
dimensional PIV is completed during Stereo PIV (once for each camera),
individual flow fields (one from each camera) will receive a value of
uncertainty at some confidence level.

Exporting Uncertainties in Insight Software
The following describes how to export the Standard and Expanded
Uncertainties for PIV.
1. In the control panel, select
the Processing tab.
2. Select an existing
processor or (Create
Figure 10-22
New…). Processing Drop-Down Menu in the
3. Select an existing Processing Control Panel
processor or (Create
New…).
4. The PIV Processor Setup dialog opens.
5. Click the Advanced>> button for more options.
6. In the PIV Advanced Settings dialog window, choose to export either or
both Uncertainty levels.
Figure 10-23
PIV Processor and PIV Advanced Settings Windows
7. Click Ok in both windows to finish.

CHAPTER 11
Programming Macros
Insight™ software provides an easy visual interface to program your

capture and experiment settings and automate it for subsequent use
through macros. Use built-in macros to automate the capture process to
compare results with multiple schemes, validate results using different
criteria, and process data using different conditions.
A processing macro stores all of the processing settings including the

Application, Spatial Calibration, Preprocessing Pipeline, Processing, and
Post-Processing Pipeline. Each processing setting and parameter, not just
the configuration names are stored in the macro file. The only settings not
recorded are the Processing Mask, calibration images used to create
calibration file, and any images used in preprocessing. Including images in
the macro would make it too large.
A macro may be loaded using the Macro drop-down menu. When loaded,
the processing configuration populates all the processing settings. The
details stored in the macro can be viewed using the same dialogs that were
used to create them.
Each macro has a name and a P#. The output files have the same P# as
the macro that created them. This allows you to recall all the processing
parameters used to create a result, even if you have tried many alternative
processing techniques.
The configuration files are copied, renamed with a date-time code, and
saved when used in a macro. The date-time code gives each configuration
a unique name, and prevents file overwriting.
Macros and configuration files with a date-time code are read only and may
not be changed. This ensures that the output files with a given P# were
created with the macro with the matching P#.
A macro or configuration with a date-time code can be used as the starting

settings for creating a new macro or configuration file. This allows an easy
and reliable method to change and document a single parameter in a
complex configuration.
The date-time code is a 12 digit number showing the month, day, year,
hour, minute, second that the macro was created in the format
mmddyyhhmmss.
The macro P# archiving scheme has replaced History Logging in previous

versions of Insight software.
11-1
When no macro is selected the output files will have P000 and the
correspondence between processing parameters and output files is left to
the user.
In addition, batch files can be used to run macros on sets of images or

other data. After creating the macros and data sets you want to analyze
using those macros, you can create a batch file that will perform these
actions sequentially.
Working with Macros

The following describes how to create macros and perform other tasks with
macros.
To Create a Macro
1. Select the processing configuration including Application, Spatial
Calibration, Processing Mask, Preprocessing, Processing, and Post-
processing.
2. From the Macro drop-down list select Save As… Enter the macro
name in the dialog box and click OK.
Figure 11-1
Enter a Name for the
Macro
3. A P# will be assigned to the macro. A date-time code will be appended

to any configuration names that do not already have a date time code.
4. To run the macro, press the Start button.
5. To create a new macro from an existing macro, load the existing
macro. Make changes to the processing configuration either by using
the Setup button to change one or more parameters, or by selecting a
different configuration, or create a new configuration. You can change
one or all configurations. When you press Start a new macro is
created with the next P#.
A macro may be renamed using the drop-down list (Save As…)
command. The macro name can be changed, the macro P# cannot be
changed.

Working with Image Lists
Image Lists define a set of images. Image lists are used in batch files to
open a set of images for processing. They can also be used to interactively
open a set of images, as an alternative to selecting them from the
RawData or Analysis folder.
To create an image list from the Exp Tree:

1. Open the RawData or
Analysis folder and select the
images.
2. Right-click and select Create
Image List.
3. Enter a name in the dialog
box. The image list is saved in
the Settings folder.
Figure 11-2
Enter Name for the New Data List
To create an image list from the toolbar:

1. Open the images using the Exp Tree.
2. Click the Create Image List button on the
toolbar.
3. Enter the image list name in the dialog. The image list is saved in the
Settings folder.
To Open an Image List:

1. In the Exp Tree navigate to the Settings folder and locate the image
list.
2. Double-click on the image list.
or right-click on the image list and select Open.
or click on the image list and while holding the mouse button down
drag the image list into image viewing area.
Programming Macros 11-3

Working with Batch Files
Batch files provide a way to run multiple macros sequentially. Since
macros are limited to analysis of a set of images within one experiment,
batch files provide a way to work across experiments. It is still the case that
one macro can only analyze images that are located within the same
experiment, but with batch files one can create multiple associations
between macros and data sets.
To Create a Batch File

1. First create the macros in each experiment in which you want to
analyze data. Refer to the previous section for instructions on how to
do this.
2. Create an image list for the set of images you want to analyze. Refer to
the previous section for instructions on creating an image list. One of
the image list options is Open Images. You could use the Exp Tree to
open the images and set the batch setup to open images instead of
creating an image list.
3. Now that the macro and image list are created, you can begin the
creation of your batch file. Click on the Batch drop-down list and select
Create New….
4. The Batch Editor dialog will open (Figure 11-3).

5. To add a process to the Batch Setups, select an Image List and Macro
and press the Add button. The Remove, Clear Steps, Move Up, and
Move Down buttons can be used to modify the Batch Steps order or
remove Batch Steps. To Include Batch Steps from another Experiment,
select the Experiment from the Select Experiment drop-down list. The
available Image Lists and Macros for that experiment will be loaded
into the Image List and Macros boxes.
6. Enter a name in the Batch Name field.

7. Click the OK button to exit and save the batch file. The batch file is
saved in the Root Folder and listed in the Experiment Tree at the
bottom of the list after the experiments.
Figure 11-3
Batch Editor
8. To Select a Batch File use the Batch drop-down list.
Figure 11-4
Batch Editor
Drop-down List
9. To change or view a batch file select it in the drop-down list.

10. Press the Setup Batch button to open the Batch Editor dialog.
Programming Macros 11-5

To Run a Batch File
1. Right-click on a batch file and select Run.
2. Or select the batch file from the Batch drop-down list. And press the
Start Batch button.
3. A batch can be stopped with the Stop Batch button.
Other Tasks
Tasks Description
Run a Batch
Click on the tool bar, or in the Tools | Visual Macro
Programmer window.
Stop a Batch
Click on the tool bar.
Load a Batch Select the batch name using the Batch drop-down menu.
Delete a Batch Use the Exp Tree to navigate to the bottom of the Tree
View, the batch files are listed after the experiment folders.
Right click on the batch file and select delete or click on the
batch file and press the delete key.

CHAPTER 12
Calibrating and
Processing PLIF
Images
Insight™ software provides a flexible platform for acquiring as well as

processing planar laser-induced fluorescence (PLIF) images. PLIF is
typically used for measurements of concentration or temperature in liquid
flows, temperature of gases using acetone, or species concentration in
combustion process. Based on the intensity captured by camera and with
the proper analysis, the result gives the proper concentration or
temperature profile of interest.
This chapter describes how to:

 Capture PLIF calibration images and,
 Process PLIF images that were captured as described in Chapter 7,
"Capturing Images".
Note: Insight software supports only 10-, 12-, and 16-bit gray scale images.
Capturing PLIF Calibration Images

Insight software provides the option of capturing PLIF images for
calibration purposes.
If you need to quantify intensities, you must capture calibration images at

known uniform concentration or temperature conditions with laser and
cameras operating with the same settings as in the current experiment or
run.
If you are processing PLIF images to determine temperature, you must

have two calibration images.
If you are processing the image with the intention of determining

concentration, a single calibration image using the origin as the location of
the second calibration point is sufficient.
12-1
If the measurement is for species in combustion process, calibration based
on the controlled combustion using the Flat Flame burner may be used.
The Flat Flame burner provides laminar flame with controls on the type and
amount of fuel to be used for the combustion. The calibration images must
be renamed and saved and moved to the settings directory in the current
experiment.
Figure 12-1
Calibration Images
To capture calibration images:

1. From Insight main menu, click Tools | Capture Perspective Cal
Images. The Capture Calibration Images box appears.
2. Click Capture.
Processing PLIF Images

Before you start processing images, make sure the following steps have
been completed:
 You have raw images captured as described in Chapter 7, "Capturing
Images".
 You have met the calibration requirements as described in Chapter 12,
"Capturing PLIF Calibration Images".
 If you need to make background correction, you have captured the
background image with no fluorescing species present in the flow but
with the laser and cameras operating at the same settings as in the
experiment, and saved in the settings folder of the current experiment.
Notes:
o If you are using two or more calibration images, the background
correction is automatic. Under most circumstances a background
image is generally not necessary.
o If you are using a single calibration image, it is recommended you
use a background image.

Setting up PLIF Processor
Before processing PLIF images, you need to set up the processing
parameters.
1. Select the PLIF Application in the processing tab. In the Processing
selection box, select Create New to create a new PLIF processor (or
select an existing processor to edit it).
2. Click Setup button, the PLIF Processor Setup dialog opens
(Figure 12-2).
3. Select the measurement type. It only affects the unit of results.
4. Select a PLIF processing method. There are two internal processing
methods (Ratiometric method and Linear Fit method) and one user-
defined method (MATLAB® Plugin). Please see PLIF Processing
Method and PLIF MATLAB Plugin for information about these
methods.
Figure 12-2
PLIF Processor Setup
Calibrating and Processing PLIF Images 12-3

5. If the PLIF images are acquired with TSI’s Laser Pulse Energy Meter,
you may check Enable Laser Pulse Energy Correction checkbox to
correct pulse energy variation. The Laser Pulse Energy Meter is to
correct for any change in pulse-to-pulse laser energy from the Nd:YAG
laser. The Energy Meter measures every laser pulse and makes the
proper correction based on the variation of the laser pulse.
6. If an absorption reference image is available, you may check Enable
Absorption Correction to correct laser absorption by using Beer-
Lamber law.
7. Edit the list of calibration images using Add and Remove buttons. To
add a calibration image, click Add button and the Add Reference
Image dialog opens (Figure 12-3).
8. Select the calibration image and the reference scalar value
(concentration or temperature, depending on the measurement type
selected). Click OK to add the reference.
Note: Up to 10 reference images can be added.
Figure 12-3
Add Reference
9. To view calibration curve, click the View Calibration Curve button.

Note: Use “Open Data Cursor” to view calibration point data
interactively.
Figure 12-4
Curve View Screen

10. Select the background image to be subtracted from the PLIF images.
Check No Background Image if there is no background image
available.
11. Select the absorption reference image, and specify the reference dye
concentration and the laser propagation direction in the image. If there
is a non-linear region in the reference image, you may choose to use a
subregion for linear fit when determine the absorption coefficient.
12. Since PLIF processor saves the scalar data as a grayscale image with
the slope and offset stored in the image header, the grayscale value
assigned to each pixel depends on what mapping method is used.
a. Use Full Range per Frame (Automatic): the scalar data will be
scaled using the minimum and maximum of the current frame.
b. User-defined range (Manual): manually enter the minimum and
maximum value and apply it to all frames processed.
13. You may click Help button to view on-line help information about PLIF
processing.
PLIF Processing Method

Insight PLIF processor performs pixel-by-pixel calibration before
processing. Therefore, the following problems that are common in PLIF
can be automatically corrected:
 Spatial variations in laser sheet intensity
 Camera noise
 Unwanted background signals.
To convert from image intensity to scalar quantities, the following PLIF

processing methods can be used:
Ratiometic Method
For each pixel in PLIF images, the two closest calibration points in terms of
pixel intensity are used to compute the scalar quantity at that pixel using a
two-point linear equation.
Linear Fit Method

For each pixel in PLIF images, all the calibration points are used to obtain
a linear equation between intensity and scalar quantity, which is then used
to compute the scalar quantity at that pixel.

PLIF MATLAB® Plugin
Besides the two built-in methods for standard PLIF processing, Insight also
allows PLIF users to define their own processing method using the
powerful and widely-used MATLAB® script through the MATLAB® Plugin.
To Set up PLIF MATLAB® Plugin

1. Select User-defined MATLAB as the processing method in PLIF
Processor Setup dialog.
2. Click MATLAB Plugin Settings button, the MATLAB® Plugin Setting
dialog opens (Figure 12-5).
Figure 12-5
MATLAB Plugin Setting
3. Select the MATLAB® script file. If the script requires command line
input, enter it in the Input textbox.
4. If you do not have the final script yet and want the data exported from
Insight software to MATLAB® to help you debug the script, check No
Script, Export Data Only checkbox.
Note: Use “Who” or “Whos” command to view exported variables in
MATLAB® command window.

5. In Data To MATLAB® list, check the data to be exported from Insight
software to MATLAB® workspace. The variables will be accessible in
MATLAB® as global variables under the name specified in this dialog
window. To edit the variable name, double-click the row to open this
dialog (Figure 12-6).
Figure 12-6
Edit Variable Name
6. In the Data From MATLAB® list, check the data to be imported by

Insight software from MATLAB® workspace. Edit the variable name if
needed.
To use PLIF MATLAB® Plugin

1. Setup the MATLAB® plugin as described above.
2. Setup the rest of PLIF processor settings like standard PLIF
processing.
3. Open PLIF images, and click Start button in the processing tab.
4. MATLAB® program will be launched in the background, and will be
closed after the script is executed.
5. The result will be imported by Insight software, displayed and saved
just like standard processing.
Note: If you have checked No Script, Export Data Only in
MATLAB® Plugin Setting dialog, you will see this message box
to notify you that the data have been exported to MATLAB®.
You can now debug your script in the MATLAB® window
opened by Insight software. It is VERY IMPORTANT not to
close the message box until you are done with the MATLAB®
program.
Figure 12-7
PLIF MATLAB Plugin Data Exported

Two Camera PLIF Processing
To process the images from two PLIF cameras, viewing the same flow
area, and combine the results into a single output scalar file, follow these
processing steps:
1. Select Application PLIF.
2. Create a Spatial Calibration using 3D Perspective Calibration.
Figure 12-8
Spatial Calibration
3. For PLIF with cameras PLIF1 and PLIF2 select Second Camera: 2.
Figure 12-9
4. Create a Perspective calibration file with CPT files from PLIF1 and
PLIF2. See Chapter 9, 2-D Calibration and Perspective Calibration for
details.
5. Process PLIF1 with Spatial Calibration: None and Processing
configured for PLIF 1 PLIF1. This processing step creates and displays
*.SCALAR1.TIF images.
6. Save the PLIF1 processing as a macro.
7. Reselect the RawData images. This is necessary because the previous
selected PLIF1 and PLIF2 image are now the Analysis SCALAR1.tif
image in the image display.

8. Configure a second macro (name it PLIF2) with:
a. Spatial Calibration selected as PLIF1 & PLIF2 calibration file.
b. Kernel Processor configured for PLIF2.
c. Postprocessor pipeline containing a Scalar Data Export processor
with the settings to “Combine with Other PLIF Data”. The Process
Number input allows the Scalar1.TIF P# to be selected.
Figure 12-10
Scalar Data
Export Setup
The Process Number input allows the Scalar1.TIF P# to be selected.
Two Camera PLIF can be processed as a batch by creating a macro for

PLIF1 and then a macro for PLIF2 with the Combine with other PLIF
Data in the post processor. Use the image list for both batch steps.

PLIF + PIV Processing
To process and combine PLIF and PIV data from two cameras viewing the
same flow area follow these steps:
1. Select Application PIV.
2. Open the Image(s).
3. Configure the processing to generate a PIV vector field as with any PIV
images using Spatial Calibration: None.
4. When you have the output vector fields you like save the processing as
a macro.
5. Select Application PLIF.
6. Create a Spatial Calibration using 3D Perspective Calibration.
Figure 12-11
Spatial Calibration
7. For Second Camera select L or Left PIV camera.
Figure 12-12
8. Create a Perspective calibration for PLIF1 and PIV Left cameras. (See
Chapter 9 for details on Perspective Calibration).
9. Select the PLIF and PIV calibration file.
10. Configure the Processing for the PLIF1 camera.

11. Configure Post-Processing | Scalar Data Export | Combine with
PIV Data.
Figure 12-13
Scalar Data
Export Setup
12. The Process Number is the P# of the PIV macro.

13. Process the Open Images. No not reselect the images.
14. The combined PLIF + PIV data is stored in the *SCALAR1.sca file.
15. Save the PLIF + PIV data combine settings as a macro.
16. To create a PLIF + PIV batch, select the images and the PIV processor
macro. Select Open Images and the PLIF macro with PIV data
combine post-processor.

Exporting PLIF Data
PLIF Processor saves the processing result as a TIFF image. To export the
PLIF data to an ASCII text file that can be loaded by Tecplot® or Excel®,
use the Scalar Data Export which is available as a post-processing step in
the PLIF application.
To Setup Scalar Data Export

1. In the processing tab, make sure PLIF Application is selected.
2. In Post-processing drop-down list, select (Create New…) to create a
new PLIF data export or select an existing one to edit.
3. Click Setup button, the Scalar Data Export setup dialog opens
(Figure 12-14).
Figure 12-14
Scalar Data
Export Setup
4. Specify the grid interval of the exported data. The pixels of inside each
grid spot in the PLIF data image will be averaged and exported.
5. Check Combine with PIV Data to export the PLIF data with the PIV
vector field with the same capture name from, e.g., simultaneous
PIV/PLIF measurement.
6. Check Combine with Other PLIF Data to export the PLIF data with
the result of another PLIF camera with the same capture name from,
e.g., dual-camera PLIF measurement.
Note 1: When combine with PIV or Other PLIF data, a perspective
calibration must be selected in the processing tab.
Note 2: When combine with PIV or Other PLIF data, PIV and PLIF
processing must be performed for other cameras before
using the PLIF data export.

CHAPTER 13
Calibrating,
Processing, and
Viewing GSV Images
Global Sizing Velocimetry (GSV) is a global imaging technique for

simultaneous size and velocity measurement of transparent and spherical
particles over a two-dimensional region. A thin laser light sheet illuminates
the particles in a plane. A camera records the out-of-focus images of the
particles in the light sheet. Particle size is obtained from the oscillation
spacing of scattered light recorded in the out-of-focus images. Particle
velocity is obtained by matching particles of same size found in two
consecutive image frames. The GSV is a technique applicable to get the
size and velocity of spherical droplets in sprays. The range of droplet size
is between 5 and 1000 microns and the concentration is not more than
3000 droplets per centimeter square.
This chapter discusses how to:

 Calibrate a GSV system
 Process GSV images to get particle size and velocity
 View GSV processing results
 View size/velocity statistics using Tecplot®
Global Sizing Velocimetry Calibration

Process
The only calibration required by GSV is the magnification before
defocusing. This is the same as PIV 2-D Calibration. The following section
describes how to perform the GSV calibration using Insight™ PIV 2-D
Calibration function. For information on how to capture calibration images,
see “Capturing Calibration Images” for details.
GSV Calibration
1. From the Processing tab of Insight software, select Calibration. The
2-D Spatial Calibration dialog opens.
2. Select Measured in the tab control box.
13-1
3. Type in the distance between any two dots in the Object size in mm
box. The standard GSV calibration target has the grid interval of 2.5
mm.
4. Click Measure.
5. Place the cursor in the center of the first dot. Click the left mouse
button once.
6. Move the cursor to the center of the second dot. Click the left mouse
button once.
7. The result is shown in Calibration box.
8. Click OK to save the calibration file for use in GSV sizing process.
Processing Global Sizing Velocimetry

Images
Insight software allows you to process GSV images to get particle size,
position, and velocity.
Processing GSV Image

1. Select one or more GSV images from the experiment tree. Drag and
drop the image(s) onto the display panel. See Chapter 6, “Viewing,
Enhancing, and Displaying Image Files” for detailed information on
how to display and enhance images for processing.
2. Select the GSV application from the Application list from the
Processing tab.
3. In the Processing group, select (Create New…) in the drop-down list or
select an existing processor.
4. Click Setup button to open GSV Processor Setup dialog.
Figure 13-1
GSV Processor Screen

5. Click Particle Sizing Settings to set up the GSV sizing processor. See
“Setting up GSV Sizing Processor” for more information.
6. Click Particle Matching Settings to set up the GSV tracking processor.
See “Setting GSV Tracking Processor” for more information.
7. Using the two checkboxes in the setup dialog, you can do:
Sizing Only Check the Skip Particle Matching checkbox.

Tracking Only Check the Use Existing Particle Data Files
checkbox. The particle data files with the
same capture name as the opened images
must be available during the processing.
Sizing and Tracking Uncheck both checkboxes.
8. Click Start to start the selected GSV processor.

9. Click Stop to quit the current GSV processor.
10. Click Process Monitor to turn on GSV process monitor for sizing
process. See “GSV Process Monitor” for more information. Note that
the process monitor is only for the sizing process.
11. GSV processing results are automatically saved to the Analysis folder
in the experiment tree. The results are also displayed on the display
panel immediately after the process. See “Viewing Global Sizing
Velocimetry Processing Results” for information on GSV data file
format and how to change the display settings.
Setting up GSV Sizing Processor

1. Click Particle Sizing Settings button in GSV Processor Setup dialog.
The GSV Sizing Processor setup dialog opens.
Figure 13-2
GSV
Processor
Sizing
2. The parameters in the sizing processor setup dialog are:
Parameters Description
Laser Camera Select the laser-camera arrangement you are using: a)
Arrangement laser on the left and camera on the right; or b) laser on
the right and camera on the left.
Oscillation Intensity Set the minimum of mean intensity of an oscillation
Threshold pattern. You may check Automatic box to use the
default value chosen by the program, or enter the
threshold (after uncheck the Automatic box).
Calibrating, Processing, and Viewing GSV Images 13-3

Relative Refractive GSV does not require the refractive index for size
Index of Particles measurement. However, if the refractive index is
known, you may enter it here (after uncheck the
Automatic box) to get the most accurate size
measurement.
3. (Optional) Click Advanced to open GSV Sizing Processor Advanced

Setup dialog (Figure 13-3).
Figure 13-3
GSV Processor
Sizing Advanced
4. (Optional) There are two groups of parameters in the sizing processor

advanced setup dialog: scan settings and oscillation settings.
Scan Settings Description

Vertical scan output Select the output of vertical scans, which can be
mean, median, or maximum of the rows included in
each vertical scan. Mean is the default choice. Median
may be a good choice if the scan window height is
large. Maximum gives the best sensitivity but it is also
the most sensitive to noise.
Vertical scan window Set the window width of windowed FFT. Check the
width Automatic box to let the program choose the optimum
value.
Vertical scan window Set the number of rows in each vertical scan. Check
height the Automatic box if the height of oscillation patterns
is between 8 and 24 pixels. Otherwise, uncheck the
box and enter a value that is about half of the pattern
height.
Horizontal scan Set the steps for windowed FFT. Larger steps results
interval in less number of windowed FFT to be calculated for
each scan, and thus faster processing. The rule of
thumb is using a value closer to the minimum
oscillation spacing you want to resolve in the images.

Oscillation Settings Description
Minimum normalized Set the minimum width of oscillation patterns relative
width to the nominal width estimated by geometric optics.
The typical range is between 0.5 and 1. Oscillation
patterns shorter than the minimum width will be
rejected.
Maximum normalized Set the maximum width of oscillation patterns relative
width to the nominal width estimated by geometric optics.
The typical range is between 1 and 1.5. Patterns
longer than the maximum width will be checked to see
if they are formed by the overlapped oscillation
patterns.
Minimum oscillation Set the minimum visibility of oscillation patterns. The
visibility I  Imin
visibility is calculated by max , where minimum
Imax  Imin
and maximum are found within each scan window.
Maximum oscillation Set the maximum contrast of oscillation patterns. The
contrast Imax
contrast is calculated by , where mean and
Imean
maximum are found within each scan window.
Maximum oscillation Set the maximum overlap percentage allowed for
overlap oscillation patterns found in adjacent vertical scans.
The overlap percentage is defined as the length of
common section over the length of whole oscillation
pattern. It is used to detect the oscillation patterns that
are processed multiple times by adjacent vertical
scans.
Minimum power Set the minimum ratio between the highest peak and
spectrum peak ratio the second highest peak in the power spectrum of an
identified oscillation pattern. This parameter is
effective in rejecting bad patterns such as those
caused by interference of nearby particles.
5. (Optional) Click OK to close the GSV Sizing Processor Advanced

Setup dialog.
6. Click OK to close the GSV Sizing Processor setup dialog.

Setting GSV Tracking Processor
1. Click Tracking Process Setup. The GSV Particle Processor setup
dialog appears.
2. Select a processor configuration from the list.
3. Click Processor Settings. The GSV Tracking Processor setup dialog
appears (Figure 13-4).
Figure 13-4
GSV Processor
Matching
4. The parameters in the tracking processor setup dialog are:
Maximum size The maximum relative size difference, in percentage,
difference allowed between two pair particles.
Minimum and The minimum and maximum horizontal shift, in pixels,
maximum between two paired particles. Moving left in the images is
horizontal shift negative shift and moving right is positive shift.
Minimum and The minimum and maximum vertical shift, in pixels,
maximum vertical between two paired particles. Moving down in the images
shift is negative shift and moving up is positive shift.
Maximum distance The maximum distance, in pixels, between neighbor
between neighbor particles. The neighbor particles are those with similar
particles motion and are used primarily in the motion match.
Maximum number The maximum number of particles selected within the
of neighbors maximum neighbor distance.
Use motion match Check the box to enable motion match, which runs after
the size match.
Maximum Particles paired by motion match should have similar shift
deviation from direction to those paired by size match. The deviation must
local flow be within the maximum deviation defined here.
direction

Maximum size Particles paired by motion match can have larger size
difference (for difference than the Acceptable size difference, but they
motion match) should still have similar size and the maximum size
difference allowed in motion match is defined here.
GSV Process Monitor

When you click Monitor button on the GSV tab, the GSV Process Monitor
dialog is displayed during processing to allow you to see what is happening
during the processing.
Figure 13-5
GSV Sizing
Monitor
The GSV process monitor has four graph panels:

1. The first panel shows the horizontal intensity profile obtained in the
current vertical scan. Depending on the output type selected in the
sizing processor setup, it can be mean, median, or maximum of the
rows included in the vertical scan.
2. The second panel shows the space-frequency distribution of power of
the intensity profile shown in the first graph.
3. The third panel shows the identified oscillation pattern with the height
indicating the oscillation frequency.
4. The last panel shows the normalized frequency spectrum of oscillation
patterns found in the current scan line.

The GSV process monitor also displays the values of oscillation intensity
threshold and scan window size, as well as the current scan line. The
process monitor also support the step processing feature which allows you
to step through the whole image one scan line at a time.
Viewing Global Sizing Velocimetry

Processing Results
Global Sizing Velocimetry (GSV) automatically saves the processing
results in the Analysis folder of the experiment tree. GSV also
automatically displays the results on top of raw images after processing.
The saved results can be viewed in a spreadsheet, or displayed on top of
the raw images. If straddled image frames are processed, the results are
displayed in both A and B frames.
Opening GSV Data File as a Table

1. Close any images opened in the display panel.
2. Select one or more GSV data file in the Analysis folder of the
experiment tree. Drag and drop the files on the display panel. Or right
click the mouse and select Open As Table from the menu.
3. The selected data file(s) is opened as a table with captions. For
particles without velocity data (unmatched or no tracking is done), a
very large number is shown in the velocity cells.
Viewing GSV Data File on Top of the Raw Image

1. Have a raw image open in the display panel.
2. Select the corresponding GSV data file in the Analysis folder of the
experiment tree. Drag and drop the files on the display panel.
3. The selected data file(s) is shown on top of the raw image. Particles
are represented by circles centered at their positions, and velocity
vectors are drawn if available.

Changing Display Settings for Particles
1. Click Tools in the toolbar menu of the display panel. Select Tools |
Particle | Setup. The GSV Particle Display Settings dialog opens
(Figure 13-6).
Figure 13-6
GSV Particle Setup
2. There are three groups of settings: particle settings for circles

representing particles, text settings for particle size text, and velocity
settings for particle velocity vectors.
Particle Settings Description

Particle Color Set the color of circles representing particles found in
the image.
Particle Scale Set the scale of circle size with respect to actual
particle diameter in pixels.
Display only matched Check the box to display only particles matched in
particles tracking process. Otherwise, all particles found in the
sizing process are shown.
Particle Style Select Outline or Filled for circles representing
particles.

Text Settings Description
Show Text Check the box to display a text next to particle showing
the particle diameter.
Text Color Set the color of text.
Text Font Set the font of text.
Vector Settings Description

Show Vectors Check the box to display vectors.
Vector Color Set the color of vectors.
Vector Scale Set the scale of vectors with respect to actual particles
displacement in pixels.
3. (Optional) Click Apply to apply new settings.

4. Click Ok to apply new settings and close the dialog.

CHAPTER 14
Processing Spray
Images
Insight™ software offers the spray analysis option to help you analyze
spray geometry. The spatial distribution of elastically scattered light is used
to evaluate spray size, spray angle, spray development, and spray
symmetry. These functions help in the development of nozzles, injectors,
and spray systems for steady or pulsed sprays.
This chapter provides a general overview of how to setup your hardware

and software to capture, process, and analyze spray images. The
information given by the spray diagnostic using these images is to provide
the patternation of the sprays, giving the concentration profile (or mass
profile) when calibration is performed. No individual size information is
resulted in this analysis.
Hardware Setup for Sprays

Spray analysis is performed using a camera, YAG laser, light sheet optics,
and optional traverse (usually providing motion control for the injector). The
camera used for spray analysis can be the Model 630090 4MP-LS camera.
The Light Sheet Generator with the laser light arm is recommended to be
used so that it can easily be rotated to produce a vertical or horizontal light
sheet, with adjustable divergence angle and waist position.
The camera may be mounted at 90-degrees side scatter, viewing a vertical

spray cross section perpendicularly (vertical light sheet). In this case, no
perspective distortion correction is needed. A relatively large camera can
be used here, because the required depth of focus is small. Thus, the laser
power can be relatively low.
For simplified viewing of both vertical and horizontal spray sections, the
camera can be mounted at a 45-degree angle to horizontal. A relatively
small camera f# is needed here, because the required depth of focus is
large. The laser power must be higher than when the camera is
perpendicular. With 45-degree viewing, all images contain perspective
distortion, but this is easily corrected by an Insight image conditioner. The
conditioner’s correction factors can be generated within Insight software
using images of a PIV calibration target. Correction factors are supplied
with Insight software for 45-degree viewing of a horizontal light sheet and
for 45-degree viewing of a vertical light sheet.
14-1
Software Setup for Steady Spray
Steady or “continuous” sprays do not vary with time (beyond the
characteristic time of the spray). Images can be captured in two ways: with
or without accompanying PIV analysis.
When PIV analysis is planned, use a thin, well-aligned light sheet for
optimum PIV processing, and to average multiple images for optimum
spray analysis. Typically averaging 10 to 25 images yields a suitably
smooth representative image for spray analysis. Steady sprays can be
considered “unchanging with time” only if the analysis time is greater than
the characteristic time of the spray, and the 5 to 10 ns YAG laser pulse
duration is much shorter than any characteristic time of current sprays.
Figure 14-1
PIV Capture Timing Setup

When accompanying PIV analysis is not planned, images can be acquired
in the Insight PLIF mode. Typically a thicker light sheet is used and “on-
chip” integration is performed by capturing multiple laser pulses with a
longer exposure time. The following figure shows an Insight timing diagram
for a 5 sec exposure time, encompassing 75 laser pulses from a single
cavity, or 150 pulses when using both cavities. To reduce background
levels, it is helpful to use a 532 nm line filter on the camera lens (TSI
P/N 610070).
Figure 14-2
PLIF Capture Timing Setup
Processing Spray Images 14-3

Software Setup for Pulsed Spray
Pulsed sprays vary greatly with time, typically having a duration of a few
milliseconds. Capture images using Insight software’s External Trigger
function and a TTL signal with a transition, sometime at or before the start
of injection. The Insight Trigger Delay setting allows you to acquire an
image at any point in the spray development process. The following figure
shows a set of three images taken at 1.7 ms, 2.2 ms, and 3.0 ms after the
External Trigger was received.
Figure 14-3
Pulsed Spray Images Captured at 1.70 ms, 2.20 ms, and 3.00 ms after the Injector
TTL Signal was Received
The full set of 12 images is shown in Figure 14-4. The External Trigger
signal usually corresponds to the logic level injector “Command” or “Drive”
signal.
When PIV analysis is planned, the strategy is to use a thin, well-aligned

light sheet for optimum PIV processing, and to mathematically average
multiple images for optimum spray analysis. Typically, averaging 10 to 25
images yields a suitably smooth representative image for spray analysis. A
sample Insight Timing Diagram for PIV mode is shown in Figure 14-5.

When accompanying PIV analysis is not planned, images can be acquired
using the Insight PLIF mode. Again, a thicker light sheet can be used and
“on-chip” integration is performed by capturing multiple laser pulses with a
longer exposure time. Images are acquired from successive injections at
the same time within the injection process, and the “average” image is built
up on the CCD. An averaged image is thus obtained at each point in time
during the injection process. The images are identified by their Delay
attribute, contained in the filename.
Figure 14-4
Insight Timing Diagram for PIV
Mode Acquisition of a Single
Steady Spray Image
Figure 14-5 shows the Timing Diagram for PIV Mode acquisition of multiple
laser pulses in a pulsed spray image.
Figure 14-5
Insight Timing Diagram for PLIF Mode Acquisition of Multiple Laser Pulses in a
Pulsed Spray Image
Processing Spray Images 14-5

Data Analysis
Statistics Insight software provides general of sprays by
calculation of statistics for user-specified Regions of
Interest (ROIs).
Intensity Profile Insight software provides general analysis of sprays
by display of intensity profiles for user-specified
locations—horizontal or vertical.
Vertical Spray Insight software provides analysis of spray angle,
spray penetration depth, and spray area.
Horizontal Spray Insight software provides analysis of spray
symmetry by sector and annulus patternation.

CHAPTER 15
Micro Particle Image
Velocimetry (MicroPIV)
The chapter explains how to use the microscope PIV system. For
information on assembly and installation of the microscope, laser, light
guide, and camera see the microscope installation manual, laser
installation manual, and light guide installation manuals.
The microscope PIV system including the light guide and microscope are
designed to be used with less than 50 mJ per pulse laser power. Higher
laser power may damage the beam attenuator, light guide, or microscope.
WARNING
Viewing unfiltered laser light though the eyepieces may cause permanent
eye damage.
The microscope PIV system comes with a filter cube. This filter cube filters
out the green 532 nm laser light (or 527 nm when high speed YLF laser is
used) from the camera and eyepieces. This is important not just for
capturing high quality fluorescence images but also for laser safety. The
microscope is designed for multiple filter cubes. If you add a filter cube that
does not filter out the green 532 nm (or 527 nm) laser light, you must
ensure that you do not look into the eyepieces with laser illumination.
Permanent eye damage may result from looking through the eyepieces at
laser light.
WARNING
The use of controls or adjustments or performance of procedures other
than those specified herein may result in hazardous radiation exposure.
WARNING
Do not look directly at the laser light emitted from the microscope
objective. Direct, Reflected or Scattered light may cause permanent
eye damage.
Laser must be operated according to these and the other instruction

manuals.
15-1
MicroPIV Overview
PIV is a technique that has been in use for many years for measuring
macro flow fields with fields of view from about 10 mm  10 mm up to 1 m 
1 m. In standard PIV a laser light sheet is used to illuminate a plane in the
fluid. The laser light sheet is thin so if the camera has adequate depth of
field, all of the particle images are in focus. Particles from outside the light
sheet are not illumined so they do not scatter light.
Using a light sheet is impracticable for most micro fluid flows. Frequently
there is no optical access from the side to introduce the light sheet. Making
the light sheet thin relative to flow channel thickness can be challenging.
Episcopic illumination with fluorescence (epi-fluorescence) was selected to
meet these challenges. Episcopic illumination is reflected light imaging
where the illumination light is transmitted through the objective.
Fluorescent particles are used to seed the fluid flow. Fluorescent particles
absorb light at one wavelength and then emit light at a longer wavelength.
A dichroic mirror and barrier filter are used to remove the illumination light
and its reflections from the particle’s fluorescent emission so that only the
fluorescent emission of the particles reaches the camera.
The scope of this manual is limited to using a microscope for PIV

measurements. For more information on the microscope and its operation,
refer to the instruction manuals from the microscope manufacturer.
Microscopy text books on microscopes are available from several
publishers. The internet has several sites devoted to microscopy. Optical
Microscopy by Michael W. Davidson and Mortimer Abramowitz gives a
good overview of modern microscope techniques.
Many microPIV systems employ low speed laser and camera (up to 30 Hz
capture frequency) because the flow fields in microchannels tend to be in a
steady state. However, there are also flow fields in microchannels
changing with time rapidly; subsequently time-resolved microPIV system
may be necessary to get high temporal resolution of the flow field. Typically
time-resolved microPIV system is made up of high speed camera and laser
(with capture rate of 1000 Hz or higher). The operation of the low-speed or
time-resolved microPIV system is essentially the same.
Microscopy Web Sites

Name URL
Molecular Expressions http://www.microscopy.fsu.edu/
How Stuff Works, http://science.howstuffworks.com/light-microscope.htm
How Light Microscopes
Work
Nikon Microscopy U http://www.microscopyu.com/
Olympus Microscopy http://www.olympusamerica.com/seg_section/seg_home.asp
Resource Center (select Micro Ed | Microscopy Resource Center)
Optical Microscopy http://www.microscopy.fsu.edu/primer/opticalmicroscopy.html
Davidson & Abramowitz

Some published papers on microPIV and microfluidics are:
Bharadwaj Rajiv, Santago Juan G., Mohammadi Bijan 2002. “Design and
Optimization of On-chip Capillary Electrophoresis,” Electrophoresis 2002,
23 2729-2744
Devansenathipathy Shankar, Santiago Juan G., Takehara Kohsei 2002.

“Particle Tracking Techniques for Electokinetic Microchannel Flows,” Anal.
Chem. 2002, 74, 3704-3713
Meinhart C. D., Wereley S. T., Santiago J. G. 1999. “PIV Measurements of

a Microchannel Flow,” Experiments in Fluids 27 (1999) 414-419 ©
Springer-Verlag 1999
Meinhart Carl D., Wereley Steve T., Santiago Juan G. 1999. “A PIV
Algorithm for Estimating Time-Averaged Velocity Fields,” Journal of Fluids
Engineering, June 2000 Vol. 122 285-289 © 2000 by ASME
Meinhart C. D., Wereley S. T., Santiago J. G. 1998. “Micron-Resolution

Velocimetry Techniques,” Developments in Laser Techniques and
Applications to Fluid Mechanics, R. J. Adrian et al. (Eds.) Springer-Verlag,
Berilin, 1998
Meinhart Carl D. and Zhang Hongsheng 2000. “The Flow Structure Inside
a Microfabricated Inkjet Printhead,” Journal of Microelectromechanical
Systems, Vol. 9, No 1, March 2000
Oddy M. H., Santiago J. G., Mikkelsen J. C. 2001. “Electrokinetic Instability

Micromixing,” Anal. Chem. 2001, 73, 5822-5832
Santiago J. G., Wereley S. T., Meinhart C. D. Beebe D. J. Adrian R. J.

1998. “A Particle Image Velocimetry System for Microfluidics,” Experiments
in Fluids 25 (1998) 316-319 © Springer-Verlag 1998
Santiago J. G. 2001. “Electroosmotic Flows in Microchannels with Finite

Internal and Pressure Forces,” Anal. Chem. 2001, 73, 2353-2365
Micro Particle Image Velocimetry 15-3

MicroPIV Operating Principles
Fluorescent Particles
Fluorescence is a process where a fluorophore, or fluorescent dye,
absorbs light at one wavelength and emits light at a longer wavelength. In
excitation a photon from a laser, or other light source, is absorbed by the
fluorophore. The absorbed energy raises the electronic state of the
fluorophore. The excited state lasts for a finite length of time, typically
1 to 10 nanoseconds. During this time the conformational changes happen
in the fluorphore which dissipate some of the absorbed energy. In emission
a photon is released returning the fluorophore to its base energy level. The
fluorophore may then repeat the excitation emission process, assuming it
has not been damaged by photobleaching.
Figure 15-1
Figure of Absorption and Emission Spectra for Duke Scientific Corporation Red
Fluorescent Microspheres
The figure shows the excitation and emission spectra for Duke Red
Fluorescent Microspheres. The excitation maximum is 542 nm and the
emission maximum is 612 nm. The particles will be illuminated with a 532
nm Nd:YAG laser where the absorption efficiency is within 10% of the 542
nm peak. The emission is from 545–780 nm.
The emission light level is affected by the fluorophore concentration,

particle size, illumination wavelength, pulse energy, pulse duration, filters,
and fluorophore environment. Most microPIV experiments will use dyed
microspheres where the concentration of the dye has been maximized.
Increasing the fluorescent particle concentration will increase the
concentration of particles in the image, but out of focus particles may
increase the background level making measurements more difficult. When
using a Q-Switched Nd:YAG laser the wavelength is fixed at 532 nm and
the pulse duration is fixed with a value of between 3 and 10 ns. The filter

cube filters have been selected to maximize the fluorescent image intensity
while removing the 532 nm laser light.
A typical Q-Switched Nd:YAG laser has a pulse duration of 3–10 ns (or

high speed YLF laser with wavelength of 527 nm and energy of up to 10
mJ). The fluorophore excitation state lasts for 1-10 ns. This means that a
fluorophore will be able to go through from 1-10 absorption—emission
cycles from a single laser pulse. This places a maximum of fluorescent
emission that can be realized from a laser pulse. Increasing the pulse
energy will increase the fluorescent emission only to a threshold level.
When more laser energy is added above this threshold, no additional
emission will be added.
Photobleaching is an irreversible damage to the fluorophore due to high-

intensity illumination. When photobleaching occurs the fluorophore is no
longer able to emit light. Usually the fluorescent particles may be run
through the flow model many times before degradation is noticeable.
Fluorescent particles should be stored in a refrigerator and not allowed to

freeze. Keeping the fluorescent particles away from light will help preserve
them.
Reference Handbook of Fluorescent Probes and Research Products,

Ninth Edition, Richard P. Haugland Molecular Probes.
www.probes.com
www.dukescientific.com
Filter Cube
A filter cube is a group of optical filters
inside the microscope that separate
scattered fluorescent light from the
illumination light. There are three light
paths that can be traced through the
filter cube. The first is the ululation
light. The second is the reflected light.
The third is the fluorescent emission
light. The reflected light is the same
wavelength as the illumination light.
The fluorescent emission is at a
longer wavelength than the
illumination light. The filter cube has
positions for three filters—exciter,
dichroic and emitter. The emitter is Figure 15-2
also known as a barrier filter. Filter Cube
The exciter is used to select the illumination wavelength for broadband light
sources. In many non-PIV epi-fluorescent applications a xenon lamp is
used as the illumination source. The excitation filter passes a narrow
wavelength range illuminating the object with a specific wavelength to
illuminate the object. Since the laser only has a single wavelength, the
excitation filter is not required for PIV.

The dichroic mirror has a coating that reflects the laser wavelength and
transmits the fluorescent emission wavelengths. For the illumination light it
is a mirror. For the fluorescent emission it is a clear glass plate. If there is
no signal received by the camera, inspect the dichroic mirror to make sure
it has the proper wavelength to separate the laser beam and the
fluorescent emission.
Figure 15-3
Epi-Fluorescent Filter Cube Light Path Illumination Light
Path
Flow Model
Illumination
Light
Objective
Dichroic
Mirror
Filter Cube
Input Laser
Barrier Filter Light
Output to
Camera or
Eyepiece
The dichroic is a mirror at the laser wavelength. The input light is reflected
by the dichroic and through the objective.

The graph below shows that the dichroic mirror has very little transmission
(high reflectivity) from 510 to 535 nm. For wavelengths above 560 nm,
more than 80% of the light is transmitted. So, for the Nd:YAG laser light the
dichroic is a mirror. Fluorescent emission with wavelengths of 560 nm and
above will pass though the dichroic mirror.
Figure 15-4
Dichroic
Wavelength vs
Transmission
Graph
The 532 nm laser light scattered off the object will be reflected back
towards the input light.
Figure 15-5
Epi-Fluorescent
Epi-FluorescentFilter
FilterCube
CubeLight
Light Path
Path Reflected Laser
Light Path
Flow Model
Flow Model
Reflected Reflected
Illumination Light
Illumination Light Objective
Objective
Dichroic
Dichroic
Mirror
Mirror
FilterCube
Filter Cube
InputLaser
Input Laser
Light
Barrier Filter
Output toto
Output
Camera oror
Camera
Eyepiece
Eyepiece

The laser light that is reflected from the experiment without fluorescence is
at the same wavelength as the input laser beam. The dichroic mirror has
high reflectivity at this wavelength and the majority of the reflected light is
reflected by the dichroic, back in the direction of the input light. A small
percentage of the reflected light will pass through the dichroic mirror. The
barrier filter blocks the light at the laser wavelength, so that it does not
reach the camera or eyepieces.
Figure 15-6
Barrier Filter
Transmission vs.
Wavelength
The emission filter is used to block the laser light and transmit fluorescent
scattered light. The emission filter has sharp cutoff to effectively block the
illumination light while passing the fluorescent emission light. This filter will
block the small amount of 532 nm illumination light that the dichroic filter
allows to pass through.
The fluorescent emission light comes from the absorption and emission by
the flourophore dye in the particles. This is the light that we want to reach
the camera. The fluorescent emission light is at a longer wavelength that
the excitation light. The dichroic has high transmission at this longer
wavelength and the light passes through with the diachronic acting like a
clear glass plate. At the barrier filter the fluorescent emission light also has
a high transmission and passes through with little loss. The fluorescent
emission light goes to the camera or eyepieces.

Figure 15-7
Epi-Fluorescent
Epi-FluorescentFilter
Filter Cube
Cube Light Path
Light Path Fluorescent
Emission Light
Path
FlowModel
Flow Model
Objective
Objective
Fluorescent
Fluorescent
ScatteringScattering
Dichroic
Mirror
Mirror
FilterCube
Filter Cube
Input Laser
Input Laser
Light
Barrier Filter
Output to
Output to
Camera or
Camera or
Eyepiece
Eyepiece
White Light Illumination

A pulsed Nd:YAG laser is the best light source for making microPIV
measurements. But with the filter cube in place an image will only be
visible if there are fluorescent particles in the field of view. For aligning an
experiment this can be a challenge. In order to see anything you must first
have the particles in the flow model, the stage position set so that the fluid
is in the field of view, and the microscope must be close enough in focus
that the particles images are visible.
It is hazardous to setup the experiment without the filter cube installed

because the laser beam will be transmitted though the eyepieces. If the
flow model has surfaces that generate specular reflections, and most do,
dangerously high light intensity levels can reach the eyepieces. Viewing
this reflection could result in permanent eye damage. Since a high intensity
laser spot is required to excite the fluorophore the reflected light will have a
dangerous intensity level.
To make setting up the experiment easier, a white light can be used as the
illumination source. With the standard filter cube, orange light can be
imaged in reflection from the flow model. The orange color is from the

wavelength range that is transmitted by the barrier filter and also is
reflected 10% and transmitted 90% by the dichroic. The flow model can be
seen even when fluorescent particles are not in place.
Magnification and Field of View

The PIV microscope system has two magnifications and field of views, one
for the eyepieces and one for the camera. The eyepiece field of view
depends on the microscope Field Number and objective power. Nikon
microscopes have a Field Number of 22. To get the imaged diameter,
divide the field number by the objective magnification. For example: for a
10x objective, the eyepiece image is 22/10 = 2.2 mm diameter. If a
magnification changer lens is installed in the microscope, multiply the
objective power by the changer magnification to get the magnification for
the denominator.
For the camera image the CCD size replaces the Field Number when
computing the field of view. A projection lens is used to relay the image
onto the CCD surface, and its magnification is multiplied by the objective
magnification to get the total magnification. The standard projection lens is
2x. The table lists the CCD sizes for several cameras.
For example, the PowerView 4MP-HS camera with a 10x objective and 2x
projection lens has a total magnification of (10*2) 20x. The field of view is
(15.15/20) 0.75mm  (15.15/20) 0.75 mm. Each pixel images a (7.4/20)
0.37 μm area.
Table 15-1
Camera CCD Sizes
Pixel Size Height
Camera μm X Pixels Y Pixels Width mm mm
PowerView 4MP-LS 5.5 2360 1776 12.98 9.77
PowerView -HP 7.4 2048 2048 15.15 15.15
PowerView 8MP 5.5 3320 2490 18.26 13.69
PowerView 16MP 5.5 4904 3280 26.97 18.04
PowerView 29MP 5.5 6600 4400 36.30 24.20

Table 15-2
Camera Fields of View for Range of Magnifications
PIVCAM
PowerView 2M PowerView 4M 14-10
Projection Field of View Field of View Field of View
Objective Lens mm mm mm
4x 1x 2.96  2.22 3.79  3.79 2.22  1.65
10x 1x 1.18  0.89 1.52  1.52 0.89  0.66
20x 1x 0.59  0.44 0.76  0.76 0.44  0.33
40x 1x 0.30  0.22 0.38  0.38 0.22  0.17
60x 1x 0.20  0.15 0.25  0.25 0.15  0.11
4x 2x 1.48  1.11 1.90  1.90 1.11  0.83
10x 2x 0.59  0.44 0.76  0.76 0.44  0.33
20x 2x 0.30  0.22 0.38  0.38 0.22  0.17
40x 2x 0.15  0.11 0.25  0.25 0.11  0.08
60x 2x 0.10  0.07 0.13  0.13 0.07  0.06
4x 5x 0.59  0.44 0.76  0.76 0.44  0.33
10x 5x 0.24  0.18 0.30  0.30 0.18  0.13
20x 5x 0.12  0.09 0.15  0.15 0.09  0.07
40x 5x 0.06  0.04 0.08  0.08 0.04  0.03
60x 5x 0.04  0.03 0.05  0.05 0.03  0.02
1X Camera Magnification 2X Camera Magnification 5X Camera Magnification
14-10 13-8
14-10 13-8
14-10 13-8
2M
4M
2M
4M
2M
4M
Eyepiece Eyepiece Eyepiece
Image Circle Image Circle Image Circle
Figure 15-8
Relationship of the Eyepiece Image Area, the Circle, and the Area Imaged by the
Camera
The eyepiece field of view is larger than the camera field of view. When
traversing the stage to align the flow model looking through the eyepieces
shows a larger area than the camera making locating the area of interest
easier. Switching to the camera view then just required a fine traverse
adjustment. When comparing what you see in the eyepieces with what you
see on the monitor, remember that only the rectangular CCD areas are
digitized.

MicroPIV Experiment Kit and Example
Experiment
This chapter gives instructions on using the TSI Model 6501100 MicroPIV
Experiment, and information on processing microPIV images. The
experiment used in the processing examples was installed on your
computer in the Experiments\Demo\uCyl folder so that you can experiment
sample images before capturing images from your own experiment.
The MicroPIV Experiment

The experiment described in this section is done using the TSI MicroPIV
Experiment kit. This kit includes a micro flow model, fluorescent particles,
syringes, tubing, and a support stand with test tube clamps. The flow is
gravity fed, or it may be driven by suction or pressure using the syringes.
The flow model contains four parallel pathways. Each flow path has several
interesting places where velocity field can be measured. Details of the
microflow model should be requested from your TSI sales representatives.
MicroPIV Experiment Setup

1. Mix Fluorescent Particle Solution. Fluorescent particles come in a 1%
to 2% solid solution. It is often desired to dilute the concentration to
reduce the background noise from out of focus particles. To dilute a 2
% solid by 10:1 with distilled water combine 1 ml of fluorescent
particles with 9 ml of distilled water. A 0.2% solids is a good starting
point. If the seeding is too dense dilute the solution more. If the
seeding is too sparse a less dilute mixture or full concentration may be
used.
2. Set flow channel on experiment holder.
3. Attach syringes to tube couplings and mount to support stand.
4. Place the dual lamp housing switch to the white light lamp position. Set
the optical path switch for the eyepiece viewing.
5. Using white light illumination focus on the area of interest. With the
filter cube in place the image in the eyepieces will be orange. This
enables viewing the flow model without fluorescent particles.
Frequently the flow channels are more visible when filled with air
instead of fluid. With fluid in the flow channel the index between the
model and fluid may be small making the channel hard to see. Using a
low power objective makes focusing and traversing the area of interest
easier.

Figure 15-9
MicroPIV Experiment Setup
6. Turn off the white light lamp. With inverted microscopes and
transparent flow models room light may be strong enough to see an
orange image through the eyepieces with both the laser and white light
lamp off. Placing the experiment cover over the flow model will block
direct room illumination. To completely block the room light, place the
experiment cover over the flow model and then place a dark cloth over
the experiment cover. The experiment cover is then used as support to
hold up the cover keeping the cloth off the model tubes.
WARNING
The laser light in the inverted microscope comes up from the objective.
Standing with your head over the microscope is dangerous. Permanent
eye damage could result. Use the experiment cover to block the laser light.
7. Switch the dual lamp housing to the laser position. Start the flow if
fluorescent particles are not already in the flow channel. Turn on the
laser pulsing using Insight software. Focus on the particles. With laser
illumination only fluorescent scattering off particles should be visible.
8. Fill input syringe with particle laden fluid.
9. Flow can be controlled by using the two syringes. To start the
experiment remove the plunger from the top syringe, and have the
plunger in the lower syringe fully inserted. Fill the top syringe with fluid
using a third syringe. Slowly pull the plunger from the bottom syringe.
The fluid will be drawn from the top syringe through the tubing and
into the flow model input reservoir, through the channels and into the
output reservoir and then into the output tube. When the output tube is
starting to draw fluid but has not reached the syringe yet, stop pulling

the plunger. Disconnect the output tube from the lower syringe.
Remove the plunger. Reconnect the output tube to the lower syringe.
The fluid should now be flowing through the flow model.
10. After the experiment has run for a while you may need to transfer the
fluid from the output syringe back to the input syringe. This can be
done by sucking the fluid from the output syringe with another syringe
with a needle to tube and transferring the fluid to the input syringe.
This method keeps a continuous fluid flow.
11. Another way to move the fluid back to the top syringe is by using the
plunger in the lower syringe to reverse the flow direction and push the
fluid back to the top syringe. Once the plunger has been pressed all
the way in, disconnect the tube from the syringe and remove or retract
the plunger from the lower syringe. Reconnect the tube to the syringe.
Press the plunger in again to push the fluid up the output tube filling
the output tube with air. Stop before the air in the tube gets all the way
to the flow model as air may affect the flow. Disconnect the syringe
from the tube, remove the plunger, and reconnect the tube. The fluid
should flow by gravity again.
12. The experiment can be run for extended periods if the syringe has
fluid, but if air is allowed into the input tube the air bubble will be
trapped and the flow might be disturbed.
13. When using pressure to drive the flow be careful not the press the
plunger too hard. With a small bore syringe and even fairly light force
high pressure can be generated. The high pressure may rupture the
flow model. Generally it is safer to pull the fluid with vacuum than with
high pressure.
14. Use bottom syringe plunger to draw fluid into the flow channel until the
fluid has been drawn into the exit reservoir.
15. Remove bottom syringe coupling to vent to atmosphere. Remove
plunger from syringe. Re-connect the tube to the syringe. Mount the
syringe to the test-tube stand to collect fluid.
16. The fluorescent particle laden fluid may be used repeatedly.
17. Switch the optical path switch to the side position to direct the light to
the camera. View the image with Insight software on the computer
monitor. If the image looks completely dark, try increasing the Insight
Lookup Table with a higher gain. You may find either pseudocolor or
grayscale gives more visible particle images.

Figure 15-10
Setting the Output Lookup Table Gain and Pseudocolor
18. Refocus and fine position the stage for desired area of interest.
19. Capture images. Look for particle displacement toggling between
frame A and frame B.
20. Process the images, Capture a sequence and use Ensemble
Correlation if necessary.
21. Process the images. Adjust the dT value so that the particles have the
optimized displacement. In many microPIV experiments the particle
seeding must be low because of background noise from out of focus
particles. In these cases the Ensemble Correlation Processing can be
used.
22. Return fluid to input syringe either by sucking with syringe or pressure
from the bottom syringe.
Ensemble Correlation Processing

Ensemble correlation processing is an analysis technique that takes
advantage of the steady flow nature of many microfluid flows while
minimizing the problems of limited seeding in microflows. Epi-fluorescent
microPIV uses volume illumination through the objective lens to excite the
fluorescent particles. The light illuminates all of the particles in the field of
view—the in focus particles as well as the out of focus particles front of and
behind the focus plane. The out of focus particles fluoresce but are a blur
instead of a sharp particle image at the image plane. The out of focus
particles add background noise but not much signal. The out of focus
particle background can limit how dense the seeding concentration may
be. Above some thresholds, adding more particles increase the number of
in focus particles but increase the background faster. The seeding
threshold is frequently lower than the 10 particle pairs for good correlation
results rule of thumb.
Ensemble correlation combines many images to create a vector field. But

instead of combining the images, the combination happens in the
correlation maps. In each realization there may be only a few or even no
particles at an interrogation spot. The few particles present will correlate

and contribute to the displacement peak. The noise peaks will also be quite
large. In the next realization the correlation map is computed for the same
input spot areas and added to the ensemble correlation map. Each particle
pair added increases the height of the displacement peak. Each non-paired
particle increases a noise peak. As the correlation maps from more images
are summed, the signal to noise ratio in the correlation peak increases.
In the ensemble processing the ensemble correlation maps are searched

for the peaks after each realization is added.
With standard PIV a 64  64 pixel spot is required to get a good vector

field.
Figure 15-11
Standard PIV
Processing with 64
64 Pixel Spot
Required for a Good
Vector Field
With 32  32 spot size processing the flow field is readily recognizable

because of the large number of bad vectors.
Figure 15-12
Ensemble
Processing Vector
Field with 1 Input
Image, Same Result
as Standard PIV
Processing Gives

With four (4) image pairs in the ensemble correlation the flow field is now
recognizable, but there are still many obviously wrong vectors.
Figure 15-13
Ensemble
Processing with Four
Image Pairs. The
Number of Bad
Vectors has been
Reduced but Many
Bad Vectors
Remain.
With 17 image pairs in the ensemble correlation the vector field looks good.
Figure 15-14
Ensemble of 17
Image Pairs no
Obvious Bad Vectors
Remain
Figure 15-15
Ensemble Vector
Field from 46 Image
Pairs

Recursive Ensemble
With recursive ensemble processing all of the images are processed with
each processing pass. On the first pass the image displacements are
measured and used to optimize the spot offsets. Then on the second
processing pass the first pass ensemble vector file is used to optimize the
spot offsets for the next processing pass. This optimization uses the
ensemble vector field that was updated after the last image in the
sequence was processed. It does not use the vector fields from individual
image pairs. After the first pass ensemble vector field is complete, a new
vector grid is generated and the images are reprocessed starting from the
first image in the sequence. The correlation maps are summed as each
image pair is added to the ensemble.
Recursive ensemble processing will be able to go to smaller spot sizes

than classic PIV processing because the effective seeding density is
multiplied by the number image pairs in the sequence.
In the example with classic PIV processing a spot size of 64  64 was

required to get a good vector field. With single pass ensemble processing a
32  32 spot size gives a good vector field. And with recursive ensemble
processing the spot size may be reduced to 16  16 pixels.
Figure 15-16
Recursive Ensemble
Vector Field with
Starting Spot Size
32  32 and Final
Spot Size 16  16

Figure 15-17
Recursive Ensemble
Velocity Field.
Starting Spot Size
32  32 Final Spot
size 16  16, Vector
Skip 2
Background Images
Background images can be used to increase the signal-to-noise ratio of
PIV images. A background image can be either a single image captured
with an unseeded flow or from a sequence of flow images, including the
sequence of images captured for making the flow measurements. Currently
there are three background images that can computed from a sequence of
images: the minimum, maximum, and average intensity images. Each of
these background image processors creates a new image of the same size
with each pixel value computed from the same pixel location from the
sequence of images. The background images are computed separately for
Frame A and Frame B images. With a two camera system the left and right
camera images are also computed separately. In the example the
background images are computed from a set of 46 image pairs. The first
image shows a typical image from the sequence. The LUT was set to linear
grayscale 16x gain for all of the images.
A solution to this is to take advantage of the steady flow nature of many

microflows. One way to do this is to add images together to increase the
seeding density. This can be done by selecting the images in the archive.
In the Insight software, select the Background Setup processor, then select
MaximumImageIntensityProcessor.
Figure 15-18
Background Processor Setup
Dialog
This will generate a new image where the pixel intensity in the maximum
intensity image is the maximum of all pixels at that location in the sequence
of images. This has the effect of increasing the seeding density without

increasing the noise much. When a particle is present its pixel intensities
will be higher than the background and its values will be used in the
maximum intensity image. The background will elevate a high background
level.
This technique can be used with image until most pixels have a particle
then the image becomes blocked and the particles lose their definition and
become blobs instead of Gaussian spots.
The main advantage of this method is that it can be used to increase the
seeding concentration for standard PIV processing.
Figure 15-19
Typical Image
Maximum Intensity Image

The maximum intensity image can be used to create a higher particle
density image. When a particle is present at a pixel the intensity will be
higher than when background fluid is there. When the maximum intensity
image from a small sequence of images is computed the images look like a
higher particle image concentration view of the flow field. When a large
number of images are used to create the maximum intensity image
particles will fill the seeded area.
Figure 15-20
Maximum Intensity
Image

Minimum and Average Intensity Image
The minimum and average intensity images are two estimates of the
background light level in the images. In the seeded area of the flow the
minimum will be when a particle is not at a pixel location. When a few
images are in the sequence, a particle may be at some pixel locations in all
of the images in the sequence. As more and more images are added to the
sequence, the particle pixels will disappear and an image of the flow
without seed particles will emerge. Particle images that remain after a large
sequence has been processed may be stationary.
With the average intensity image as a small number of images are

averaged particle images will still be visible. As more images are averaged
together the particles in the flow will average out.
The average intensity in a higher value will be removed in background

subtraction. In the flow area some of the particle intensity will be removed
because the average will be the average seeding concentration time the
particle intensity higher than the minimum. The minimum will not remove
any of the particle image intensity but will allow some noise in the image.
Both the minimum and average images are effective at finding stationary
particles, walls, and scratches and smudges in the flow model. In these
example images the flow model had been used several times before this
experiment and the minimum and average images show stationary
particles and smudges.
Figure 15-21
Minimum Intensity
Image

Figure 15-22
Average Intensity
Image
Figure 15-23
Typical Image with
Minimum Intensity
Subtracted. The
Linear Gain has
Been Increased to
32
Figure 15-24
Ensemble Vector
Field 1 Image Pair
with Minimum
Intensity Image
Subtracted. The LUT
Gain has been
Increased from 16 to
32

Figure 15-25
Ensemble Vector
Field with 9 Image
Pairs no Obvious
Bad Vectors
Remain. Without
Background
Subtraction 17
Image Pairs were
Required to get a
Similar Quality
Vector Field
Figure 15-26
Ensemble Vector
Field 46 Image Pairs
with the Minimum
Intensity Image
Subtracted
Figure 15-27
Ensemble Vector
Field 46 Image Pairs
with Minimum Image
Subtracted. There
are no Holes in the
Vector Field

Figure 15-28
Ensemble Vector
Field 46 FROM
Image Pairs. There
are 9 Holes in the
Vector Field at
Locations where
Fixed Particles
Appear in the
Minimum Intensity
Image

CHAPTER 16
Presenting Data with
Tecplot® Software
Insight™ software provides Tecplot, a third-party software to display and

manage PIV data. TSI Tecplot® Add-On is an add-on software used to view
Insight software 2-D and 3-D vector files and compute flow properties, it is
also used to view Insight GSV particle files and compute diameter
statistics. Tecplot® software is a general-purpose data viewing package,
with an emphasis on fluid flow data. It offers comprehensive fluid flow data
visualization with many options. It provides the most frequently-used PIV
and GSV flow field viewing controls in a single tabbed dialog. The
combination of common PIV and GSV data viewing using TSI Tecplot®
Add-On and the ability to customize the graphs to meet your requirements
using Tecplot® software makes this feature a powerful tool for analyzing
your PIV and GSV data.
Installing Tecplot® Software

TSI Tecplot® Add-On is installed with Insight software. However, Tecplot®
software has to be installed separately. Follow the installation instructions
provided with the Tecplot® manual.
If you have not installed Insight software into the default folder C:\Program
Files\TSI\Insight 4G\, you would need to set the Insight folder using the
Msg tab option. See “Message Folder.”
Tecplot is a registered trademark of Tecplot, Inc.
16-1
Accessing and Using TecPIV and Tecplot ®
Software
To open Tecplot® program:
Click Tecplot on the Insight Toolbar. The Tecplot® program opens and the
TecPIV dialog becomes available. The following screen shows the different
options available to display and manipulate the 2-D, 3-D, and particle data
generated by Insight software. Following are descriptions of the tasks you
can perform with each of the TecPIV display features and options. You can
configure Tecplot® software to directly open a processed image by
configuring the Msg tab option.
Figure 16-1
TecPIV Dialog
Screen
Selecting and Reading Files into Tecplot®

Software
Use the File tab options in the TSI Tecplot® Add-On dialog to select and
read vector and particle files into Tecplot® software.
To Select and Open Vector Files

1. From the TSI Tecplot® Add-On dialog, click File | Select Files.
2. The Select Vector File(s) dialog opens. Use the Look In scroll box to
navigate to your vectors folder. Select All Vector Files in the File of
type box to view the vector files, or select “All Particle Files” in the File
of type box to view the particle files.
3. Select the desired vector files and click Open. The group of files is
checked for compatibility when the files are read into Tecplot®
software. To be compatible, the vector files must be the same type,
have the same number of vectors, and the vectors must be at the
same locations. If this is not true, an error message explaining why the
files are not compatible is displayed. If you try to work with
incompatible files, the results are unpredictable and a crash is likely.

To open a single vector or particle file:
 Double-click the filename, or highlight the file and click Open.
 The number of files selected is displayed in Total Number of files
of box. This number of files increases by 1 if you create a vector
field average.
To Select More Than One File

1. Hold the Control key down and click on all the filenames.
2. Click Open.
or
1. Select a group of files, highlight the first filename.
2. Hold <Ctrl>+<Shift> and click on the last filename.
3. Click Open.
The following dialog box will show if you choose vector files:
Figure 16-2
Vector File
Dialog Box
The following dialog box will show if you choose particle files:
Figure 16-3
Particle File
Dialog Box
Presenting Data with Tecplot® Software 16-3

The following dialog box will show if you choose scalar files:
Figure 16-4
Scalar File Dialog
Box
Auto Binary Conversion

Insight software generates ASCII text files. When Tecplot opens an ASCII
text file it converts it to binary in RAM while loading the data. The
converting to binary can take quite a bit of time when hundreds or
thousands of files are being opened. To make the file loading process
more efficient, the ASCII files can be converted to binary and saved on disk
before the data is loaded into Tecplot. The data conversion and saving
takes a little more time than opening the files from ASCII. Once the files
have been converted, the time to load the data again is much shorter.
Figure 16-5
File Selection
Dialog with Auto
Binary
Conversion and
File Converted to
Binary OK
Message
The binary versions of the files are saved with the same name as the
ASCII version, with a “B” pre-pended to the file extension.
ASCII File Extension Binary File Extension

VEC BVEC
V3D BV3C
SCA BSCA
PAR BPAR
When the Auto Binary Conversion is checked the file loader first checks to
see if a binary file with the same name is available and loads the binary file

if available. Only the ASCII file extension files are listed in the file selection
dialog. The binary files are selected by choosing the associated ASCII file.
If the binary file is not found it is created with the Tecplot DOS application
PrePlot. When converting a sequence of images you will see a cascade of
PrePlot DOS windows opening and closing.
The message under the Auto Binary Conversion Check gives the status of
the binary files. If there is an error in converting a sequence of files to
binary, the binary conversion will be turned off and the files loaded from the
ASCII text files, and the status text set to inform you of the problem.
P000 File Caution

The file P Number is set by the macro. If no macro is used, the P Number
is P000. If you have created P000 files and converted them to binary,
change the processing settings, reprocess, and open the new files. The old
binary files will be loaded into Tecplot not the just processed ASCII file
data. To load the new data either unselect the Auto Binary Conversion
check or delete the old binary files using Windows® Explorer browser.
With P Numbers, other than P000, the processing for a P Number is fixed
and the reprocessed ASCII files will be the same as the old ASCII files and
it is not necessary to keep track of when the binary files were converted.
The following are other options available to open and display vector files.
Option Description
Previous, Next Use Next and Previous to change the displayed file by one
in the file list. If you are already at the end of the list, the
display file is not changed. To go directly to a specific file,
open the File pull down list and select the file or type in a
new number in the file number edit box. The display file
number is updated when you change display file.
All Use to display all open vector or particle fields together.
Seeing all of the vectors or particles at a location shows the
range of velocities at that point. When All is selected, the
filename displayed on the graph changes to the experiment
family name. When the All option is selected, the statistics
page shows the statistics for all open vector or particle
fields, including the average field if it has been created.
Showing contours as colored arrows generally gives a better
result with All selected. Reducing the number of displayed
vectors with the Arrows page skip vectors may improve the
display
Animate Select to show all of the vector or particle fields in order.
Display starts with the first field and stops when the last field
is displayed. This Animate command updates the filename
on the graph; the standard Tecplot® Animate command
does not.
To record animations use the Tecplot Main Menu | Tools |
Animate | Zones dialog.

Selecting the Plot Types
The Plot Type option is used to select the type of plot to display vectors
and particles.
Figure 16-6
Plot Type Screen
Examples of Different Plots

Arrow/contour plot for vector and particle files:
Figure 16-7
Vector Plot for Vector and Particle Files

Figure 16-8
Vector/Scatter Plot for Vector and Particle Files
Scatter plot for vector and particle files:
Figure 16-9
Scatter Plot for Vector and Particle Files

Histogram plot for vector and particle files:
Figure 16-10
Histogram Plot for Vector and Particle Files
Changing the Shape and Size of Arrows

The Arrow options are used to manipulate the shape and size of the arrows
representing the velocity vectors.
Figure 16-11
Arrow Option
Screen
To Change the Length of the Arrows

1. Click Arrow in the TSI Tecplot® Add-On dialog box.
2. Use any of the following options:
 Click Shorter or Longer to make the arrows shorter or longer than
what is displayed. The arrow scale factor increases or decreases
by 2.
 Type in a scale factor directly into the Arrow Scale box.

 Click Auto Scale to set the arrow scale to an optimized arrow
scale based on the number of vectors and the maximum velocity.
To toggle arrowhead length from fixed to a scale proportionate to arrow
length:
 Click Head by Length or Fixed Head.
To control number of vectors to be displayed:
 Provide numbers in the Skip Vectors box. A skip value of 1 means
show every vector, skip 2 means show every other vector.
To improve visualization you may want to show a reduced vector

field overlaid on flood contour field, like vorticity. When you do this,
the contour values are computed at full resolution, but only some
of the velocity vectors are shown. Using a vector skip in the
direction of the flow may enhance your ability to see changes in
the flow profile as shown in the following figure. The following
figure shows the Vector field display using X Skip = 4, Y Skip = 1
to enhance the visualization of the flow profile with x position.
Figure 16-12
Vector Field Display Screen

Selecting Velocity Vectors and Modified
Vectors
Modified vectors are velocity vectors that have been modified in some way.
If you have subtracted the mean velocity from the flow field, by toggling
between the velocity vector and modified vectors, you may view the flow
field from two perspectives. A modified vector field must have been
computed to view the modified vectors. See “Subtracting Vector Files” for
details.
Selection Description
Velocity Vectors To have velocity values read from the vector file.
Modified Vectors To have modified velocity values read.
Setting and Changing Contour Properties and

Color
The color tab options in TSI Tecplot® Add-On allow you to control contour
parameters and to set the color of the arrow displays. The following
provides detailed descriptions of the color options for vector fields and
scalar fields.
If you choose vector files, the following dialog box will show:
Figure 16-13
Color Options
Screen

If you choose scalar files, the following dialog box will show:
Figure 16-14
Scalar Contour
Screen
Selecting Contour Property

This option allows you to select a contour parameter to be displayed. The
properties lists are different for 2-D and 3-D vector fields. This property is
the variable that is used for contour values and arrow color.
To Select a Contour Property

1. Click Color tab in the TSI Tecplot® Add-On dialog box.
2. From the Property pull-down list, select a property. See “Table of Flow
Properties” for a list of properties.
Table of Flow Properties for Vector Fields

Property File Type Description
U All U velocity component
V All V velocity component
W 3-D W velocity component
CHC/Count All Vector choice code for a single vector field, the
number valid vectors at the node for an Avg field.
Standard Uncertainty 2-D The standard level (68%) of predicted uncertainty of
measurement for 2-D PIV.
Expanded Uncertainty 2-D The standard level (95%) of predicted uncertainty of
measurement for 2-D PIV.
Velocity Magnitude All Flow velocity
Vorticity All Formula
DU/DY All Formula
DV/DX All Formula
DU/DX All Formula
DV_DY All Formula
Q All Formula
Lambda_Cl All Formula
-u’v’ Reynolds Stress Sequence See Average Page | Compute Reynolds
Stress/Density button
-u’w’ Reynolds Stress 3-D Sequence See Average Page | Compute Reynolds

-v’w’ Reynolds Stress 3-D Sequence See Average Page | Compute Reynolds
Residual 3-D Residual error when computing the 3-D velocity from
the left and right vectors using least squares fit. The
error is the square root of the X, and Y errors, in
pixels
Total Turbulence Sequence Turbulence Velocity magnitude turbulence
Computed
U Turbulence Sequence Turbulence U velocity component turbulence
Computed
V Turbulence Sequence Turbulence U velocity component turbulence
Computed
V Turbulence Sequence Turbulences U velocity component turbulence
Computed
Table of Flow Properties for Scalar Fields

Scalar property All The scalar data.
CHC/Count All Scalar choice code.
Selecting Contour Type

This option allows you to select background contour.
To Select a Contour Type

1. Click Color in the TSI Tecplot® Add-On box.
2. Select a property from the pull-down list. See Table of Contour Types
for a list of contour types.
Table of Contour Types

Contour Type File Type Description
None All Background contour off
Flood All Contour shown as filled colors
Color Line All Color coded lines show the divisions
between contour thresholds
Flood + Line All Flood contour with the contour divisions
highlighted with black lines
Black Line All Contour shown as black lines, good for
laser printing
Carpet Plot 3-D Mesh connecting the vector heads. Shows
a surface with 3-D deformation of the flow.
®
To turn off the mesh using the Tecplot
Mesh check box, under File on the main
menu.

Selecting Vector Fields Arrow Colors
Select color codes to go with contour parameter values for the arrows.
Using colored arrows is effective for showing a flow parameter with high-
resolution vector fields, where adding a background contour makes the
graph too cluttered.
Use black arrows to show the velocity field overlaid on a contour

background.
To Select Arrow Colors

1. Click Color tab in the TSI Tecplot® Add-On box.
2. Select a color from the Arrow Color pull-down list. Select Off to turn off
the arrow display. This is a useful setting for viewing the contour
background.
Selecting Number of Colors

Select number of colors to represent the number of contour threshold
values in the graph. Using a high number up to 151, gives the smoothest
flood contour plots and colored arrows. Using few contours gives better-
looking line contour plots. The display speed slows down with more colors.
To Select Number of Colors

2. Type in a number in the Number of Colors box for vector fields or
Count box for scalar fields.
3. For Vector Fields, click Color Scale to select the range of colors used
in the plot. The range options change with the parameter shown.
Properties that can be either positive or negative, like U, V, W,
Vorticity, dU/dy, … can be scaled with Symmetric where the maximum
value = -1 x minimum value. With symmetric, the contours are
symmetric about 0 and 0 is always green. Range sets the minimum
and maximum contour values to the minimum and maximum values in
the data. 0 is not always in the center of the graph.
Properties that are always positive like Velocity Magnitude and

Turbulence can be shown with a color scale from 0 to maximum, or the
actual velocity range.
CHC/Count is an integer field and does not allow scaling options.
Selecting Contour Color Range

The contour color range can be selected by choosing one of four options.
Note: For Each Field, All to Display, All Fields settings, the contour levels
are automatically set according the contour parameter range for one
or all vector fields.

To Select the Contour Color Range
2. Select one of the following check boxes:
Field Description
Each Field Select to set the contours for each field as it is
displayed. When you animate a sequence of fields, the
colors are set so that each field has the full range of
colors. This is helpful in seeing the pattern in each field,
but makes comparing the values between fields more
difficult.
All to Display Select to set the contour levels according to the range
of current display field. The same range is used for all
fields during animate.
All Fields Select to set the contour range to the range of all open
fields. The same range is used for all vector fields
during animate.
Manual Enter a value for setting the contour levels in the box
below.
Selecting the Scalar Contour Minimum and Maximum

Enter value into Minimum or Maximum box and either leave the edit box or
press the apply button.
Selecting a Background
Use this option to toggle the display background color between black and
white. The black background is usually preferable when displaying on the
computer monitor while the white background works well for printing.
To toggle background color

1. Click Color in the TSI Tecplot® Add-On box.
2. Click Black Bkgrnd or White.

Computing Average Fields and Other Quantities
The Average set of options are used when average vector field and other
quantities from the Insight vector files need to be calculated. The following
describes the features available in this set.
Figure 16-15
Average Option
Screen
Computing Average Fields

This option allows you to compute the average field which is the mean of
the valid velocity vectors at each point. The number of valid vectors is
stored in the CHC (choice) data column. (See Appendix A, “Insight Data
Files”). The average contour field is the average of the contour values at
each location. This is not the contour of the average vector field. The
average field can be subtracted from each individual vector field provided
in the subtract tab option. See “Average Vector (*AVG.VEC) File” for more
information.
To Compute Average Fields

1. Open all vector fields to be included. See “Selecting and Reading Files
into Tecplot Software.”
2. Click Avg. tab in the TSI Tecplot® Add-On dialog box.
3. Select Compute Average Field.
4. Enable the Save Turb checkbox to output the standard deviations of
the velocities. See Mean and Standard Deviation (*.STD) Vector File.
5. Click Save.
Computing Reynolds Stress/Density

This option allow you to compute the Reynolds Stress for a 2-D vector
field, the average velocity field is computed and appended to the vector
field list. The average only includes the valid vectors so each node may
have a different number of vectors. This gives the average u and v velocity
at each vector node. The average velocity field is subtracted from each of
the instantaneous vectors giving u’ and v’ for each vector. The -u’ and v’
components for each instantaneous velocity vector are multiplied giving -
u’v’. The -u’v’ values are averaged at each node giving the Reynolds

Stress for the sequence of vector fields. When the Reynolds Stress is
shown as the contour parameter, the sequence Reynolds Stress is shown
as the contour parameter for each of the instantaneous vector fields.
When 3-D vector fields are used, the average W and w’ are also computed
and the Reynolds Stress values -u’w’ and -v’w’ are computed.
Reynolds Stress includes a density term that is not included in the Tecplot®
equation. To compute the actual numerical value of Reynolds Stress,
multiply the compute value of Reynolds Stress/Density by the density.
Note: The Reynolds Stress -u’v’, -u’w’ and -v’w’ values are not normalized.
To Compute Reynolds Stress/Density

3. Select Compute Reynolds Stress/Density.
4. Click Save.
Computing Turbulence Intensity

This option allows you to make turbulence computations. If it does not
already exist, it creates an average velocity field, then computes the
turbulence intensity using the formula specified. The Total Turbulence
intensity is displayed when the computations are complete.
To Compute Turbulence Intensity

3. Make the following selections:
Option Description
Standard Select this check box if the turbulence is to be computed by
Deviation the standard deviation of the velocity vectors at the vector
node location. The Standard deviation is divided by a
velocity value in the other two turbulence formulas.
Node Select this check box if the standard deviation of the velocity
Average values is divided by the average velocity at the node. Each
vector uses the local average velocity with node average.
With Node Average a high turbulence reading may be
caused by either a wide range of velocity values at the point
or by a low average velocity value. The Max Turbulence is
used to clip the maximum turbulence computed.

Option Description
Single U, V, With Single U, V, W, the velocity standard deviation at each
W node is divided by the same velocity value. You may set this
velocity to divide by any positive value you want. The
velocity used is shown to the right of the Single U, V, W
Check box.
Single U, V, U, V, W
W (cont.) The velocity value used for Single U, V, W Turbulence.
Enter velocity values by editing the numbers, or by clicking
Set Mean or Set Max. The velocity components must be
positive. You may only edit the W velocity component with
3-D vector fields. The turbulence values are clipped to this
maximum value. This is most useful with Node
Average Turbulence where the divisor may be near zero.
Set Mean
Fills in the mean U, V, W (for 3-D vector fields), of the
Average vector field into the Velocity U, V, W boxes in the
Single U, V, W Turbulence computation.
Set Max
Fills in the maximum U, V, W (for 3-D vector fields) of the
average vector field and not the velocity U, V, W edit boxes
used in the Single U, V, W Turbulence computation.
4. Click Compute Turbulence Intensity.

5. Click Save. This saves the average vector field both with and without
turbulence information to disk. A sequence of vector fields must be
open for the averaging to work. If 2-D *.vec files are open,
familyAvg.vec and familyAvg.std are stored to disk. If 3-D *.v3d vector
files are open, the average fields are saved as familyAvg.v3d and
familyAvg.v3s, when “family” is the name entered in a file save dialog.
The average *.vec and *.v3d files have the same file format as the
individual vector files so they may be opened as part of sequence with
individual vector files. The *.std and *.v3s include the velocity standard
deviation for each vector node. This allows you to view the turbulence
information without having to open up the full sequence of files. Since
the file formats are different, the *.std and *.v3s files may not be
opened with a sequence of *.vec or *.v3d files.

Subtracting Vector Files
The Subtract tab options allow you to manipulate the vector fields.
Figure 16-16
Subtract Option
Screen
The following provides detailed descriptions of these options.
Subtracting Reference Field

The Subtract Reference Field button subtracts the selected reference
vector field from each open vector field. If you have computed an average
vector field, it is used as the default reference field. Select any open vector
field as the reference field by selecting it from the pull-down list. The
subtract reference field subtracts the reference velocity from the vector
field velocity on a vector node by vector node basis.
Subtracting the average velocity field breaks the vector fields up into the
average velocity, average velocity value, and fluctuating velocity, the
individual vector field—the average vector field.
The graph title shows the name of the reference vector field that has been
subtracted.
To Subtract Reference Field

2. Click Sub. tab in the TSI Tecplot® Add-On dialog box.
3. Select Subtract Mean Velocity or Subtract Vec.
Subtract Mean Velocity

Computes the mean velocity of the currently displayed vector field and
subtracts that velocity vector from each vector in each vector field. The
graph title shows the velocity value subtracted.

Subtract Vec
Subtracts the user input U, V, W vector from each vector in each open
vector field.
Statistics Options
The Statistics tab has options to show the velocity component and contour
parameter mean, standard deviation, minimum and maximum values of the
currently displayed vector field. If all fields are being displayed, the
statistics are for all of the open vector fields. The contour statistics are
shown for the currently displayed contour parameter.
Figure 16-17
Statistics Option
Screen
Velocity/Modified Velocity Check

Look at the statistics of either the actual velocity values or the modified
velocity values. To look at the statistics of the modified vectors, if you have
subtracted the mean vector field, the modified vector field gives information
about the variations from the mean.
CHC Statistics Button

Opens the vector editing statistics dialog box.

CHC Choice Code Statistics Dialog
The Choice Code Statistics show the source of the velocity vectors.
Measured vectors are vectors directly from the processing of the image.
Interpolated vectors are places where the surrounding vectors were
interpolated to fill in the holes. Temp Blank means the vector has been
removed by vector editing and could be filled in with an interpolated vector.
Other includes points removed from the vector field by Polygon Edit, SNR
failure vectors and Smoothed vectors. The Choice Code Statistics Dialog
screen is shown below.
Figure 16-18
Choice Code Statistics
Screen
Stream Option
A stream trace shows the hypothetical path of a particle in the vector field
assuming the velocity at each point remained constant. The figure below
shows the Stream trace tab options.
Figure 16-19
Stream Trace
Screen
Number of Stream Traces

Each time a rake position button is pressed, that many stream traces are
added to the display. If 30 stream traces are selected and you press Left
and then Right, a total of 60 stream traces are shown.
Clear
Erases the stream traces from the graph.

Rake Position
The rake is a line divided with the Number of Stream traces that are the
starting position of the stream traces. The stream traces are equally
spaced at the rake position. The stream traces are computing in both
forward and reverse directions. By selecting a rake position you can see
how the flow moves relative to the location. The left, right, bottom, and top
rakes are positioned a small distance from the edges.
View Option
The view page is used with 3-D vector fields only. It has a set of seven
preset viewing angles. Switching view positions can help in visualizing the
third velocity component. The Tecplot Main Menu | View | 3D Rotate
dialog can be used for full control of the viewing perspective. The View
Page screen is shown below.
Figure 16-20
View Page
Screen

Changing the Shape and Size of Particles
The particle options are used to manipulate the shape and size of the
points representing the particle sizes.
Figure 16-21
Vector/Scatter Graph
Figure 16-22
Particle Size
Screen
To Change the Size of the Points

1. Click Particle in the TSI Tecplot® Add-On dialog box.
2. Use any of the following options:
 Click Shorter or Longer to make the arrows shorter or longer than
what is displayed. The arrow scale factor increases or decreases
by √ .
 Type in a scale factor directly into the Arrow Scale box.
 Click Auto Scale to set the arrow scale to an optimized arrow
scale based on the number of vectors and the maximum velocity.

To Change the Shape of the Points
1. Click Particle in the TSI Tecplot® Add-On dialog box.
 Select the shape from the Shape Type list. Check Particle Fill to
fill colors within the shape.
Viewing the Statistics of Particles

The Stat options are used to display the statistics of particle sizes.
Figure 16-23
Viewing Statistics
of Particles
Screen
To View the Statistics of Particle Statistics

1. Click Stat in the TSI Tecplot® Add-On dialog box.
Message Option
The Message page shows the folders of the programs that TecPIV
communicates with. To change from the default folders, press the Edit
Folders button. The Message Count shows how many messages from
Insight software have been received, the Ticks field increments each time
TecPIV checks for a new message from Insight software. When a message
is received, it is shown in the box above the message count. The figure
below shows the Message Page screen.
Figure 16-24
Message Option
Screen

Auto Open Vector Files Check Boxes
Use these check boxes to turn on or off the types of files Tecplot®
automatically displays from Insight software. Use the Edit Folders button to
have Tecplot® remember the selections. To view V3D stereo 3-D vector
fields but not the 2-D *.VEC files, uncheck the *.VEC box and check the
*.V3D then Tecplot® will only auto display the 3-D fields. To temporarily
turn off the auto display so that you can change the display parameters
without being interrupted by reading new vector fields, uncheck the boxes,
go to the other pages and set the display parameters you want, then check
the boxes and have the new vector fields displayed with your desired view.
Figure 16-25
Folders Screen
Edit Folders Dialog

The Edit folders dialog shows the locations where TecPIV will look for files.
Experiment Folder
The Experiment Folder is the directory that is the starting point the first time
you press the Select Files button in the file page. When this is the root
folder where your PIV vector files are located, you will have less navigation
to do to open vector files. When a file has been opened it that folder, it
becomes the starting location for the next Select Files.
Message Folder
Insight software communicates with TecPIV by writing a file named
Message.txt into this folder. This folder must be set to the location that
Insight software is writing messages into for the messages to be received.
Preplot Path Name

This is the full path name to the file Preplot.exe. Preplot is part of the
Tecplot® package and is usually in the Tecplot\Bin\ folder. If you are not
using the latest version of Tecplot® software you will probably have to

change this to your Tecplot\Bin folder. Preplot is used to convert Insight
ASCII files into binary files that can be opened faster by Tecplot® software.
Save Button
The Save button writes the folder information into a TecPIV.INI initialization
file. Each time TecPIV is launched, it checks to see if TecPIV.INI is
available. If it is, your stored values are automatically loaded. If TecPIV.INI
is not found, the default values are shown. If the TecPIV.INI file format has
changed since you saved it, a warning box showing the default values will
be used. The Edit Folders Dialog screen is shown below.
Time-Resolved Toolbox
The Time-Resolved Toolbox consists of two tabs in FocusPIV: Time and
Tracking.
 The Time tab eases the navigation and animation (including exporting
of movies) of many 2-D and 3-D vector files. It also provides a way to
quickly create time-history plots and calculate spectral analysis for
different variables.
 The Tracking tab enables you to track the minimum/maximum value of
any variable over time.
The following sections provide detailed information for both tabs.
Time
1. In FocusPIV, activate the Time tab. The Time dialog box opens.
Figure 16-26
Tracking Dialog
Box in FocusPIV
2. Each vector file will require a timestamp. If the files do not have a
timestamp, or you want to re-write a timestamp, enable the Overwrite
Timestamp check-box.
3. Click the button, Enable Time Analysis. This will write a new
timestamp.

The following table details each option in the Time tab.
Option Description
Overwrite Timestamp If enabled, clicking Enable Time Analysis will re-write a
Tecplot® compatible timestamp to each file.
Initial The value assigned to the first file’s timestamp.
Delta T The increment value added to Initial for subsequent file
timestamps.
Start Time The timestamp of the first file to be displayed during
animation.
End Time The timestamp of the last file to be displayed during
animation.
Go To Time When pressed, will navigate to the file with the specified
timestamp.
Time Step Skip The stepping size used while navigating between files.
Animate Speed Frames per second used during animation.
Animation Proceeds:
Forward From Start Time to End Time
Backward From End Time to Start Time
Loop Forwards* From Start Time to End Time in a loop.
Loop Backwards* From End Time to Start Time in a loop.
*Click Stop in the
Tecplot® Status Bar to
end animation.
Export Exports an AVI format file of the animation.
Time Analysis Probe When selected, move the cross-hairs onto the flow field
Discrete Fourier to obtain a variable-specific time history and a spectral
Transform Options: analysis of the signal.
Rectangular Transforms the time-history signal into the frequency
Triangular domain. The timestamp is the independent variable.
Hann Filters applied to the signal prior to the FFT to taper the
Hamming data.
Navigation toolbar. From left: Back to Min. Back one file.
Play animation on screen (at Animate Speed and with
Time Step Skip). Forward one file. Forward to Max.
Tracking
The tracking tool enables you to record locations (via scatter plot) where
maximum and/or minimum values of any property found in a specific
Tecplot® zone. This can be particularly useful for vortex tracking, where
you might want to track max/min values of vorticity, Q Criterion, or .
1. In the Color tab, choose the property you want to track.
2. For tracking, each vector file will require a timestamp. If the files do not
have a timestamp, or you want to re-write a timestamp, enable the
Overwrite Timestamp check-box in the Time tab. Click the button,
Enable Time Analysis to write a new timestamp.

3. Activate the Tracking tab. The Tracking dialog box opens.
Figure 16-27
Tracking Dialog
Box in FocusPIV
The following table details each option in the Tracking tab.

Option Description
Create Tracking Locates maximum/minimum values of the property and
marks the locations as a scatter plot. Processes all
timestamps at once. A new tracking zone is created.
Animate Tracking Locates maximum/minimum values of the property and
marks the locations as a scatter plot. Processes and
displays each timestamp at a time. A new tracking zone is
created.
Delete Tracking Deletes the tracking zone.
Overlay/Hide Tracking Toggle between showing/hiding the tracking zone.
Track Absolute Value If enabled, the tool tracks the absolute value of the
(Min/Max) property. If disabled, only the maximum value is recorded
in the tracking zone.
Track Multiple If enabled, any location at a particular timestamp that has
Vortices a property value greater than the Set Threshold value is
recorded in the tracking zone.
Set Threshold The value at which Track Multiple Vortices will threshold.
Note: You must click the button, Set Threshold, to
apply the value.
Apply Radius When tracking multiple vortices, only one max/min
property value will be plotted inside a circle with this
radius.
Note: You must click the button, Apply Radius, to apply
the value.
Mark Maxima/Clear When clicked, this will place a circle (with a radius
Marker(s) specified by Apply Radius) centered around the max/min
property value. When tracking multiple vortices, only one
max/min value will be recorded inside this radius. Click
Delete Tracking to initially mark the maxima.
Color Dropdown Change the color of the scatter plot markers.
X, Y, Z & Value The coordinates and magnitude of the maximum property
value for a particular timestamp.

Histogram Plot
The data can be displayed as a histogram by selecting the Histogram Plot
radio button in the Plot Type Page. A data file must be open for the Plot
Type Tab to be displayed. The plot files are selected on the file page. To
view the histogram from a single data file, select the file with the Forward
and Next buttons. The histogram of multiple data files can viewed by
selecting the All button. The following screen shows the plot type page
with histogram plot selected.
Figure 16-28
Selecting
Histogram Plot

The histogram is only displayed when the Histogram page is selected.
When another page is selected, the display goes back to the Arrow and
Contour or Scatter plot. Selecting the Histogram page again returns the
histogram graph. The following screen shows a histogram plot with
cumulator enabled.
Figure 16-29
Histogram Plot with Cumulator Enabled
Y-Axis Parameter Selection

The parameter to plot is selected from the Y-Axis Parameter pull-down
menu.
Figure 16-30
Y-axis Parameter Selection

Bin Width and Number of Bins
The histogram Bin Width and Number of bins are edited by entering a
value in the edit boxes. The entered value takes affect when you select
another parameter to edit or press the Apply button. When a Bin Width or
Number of Bins value is changed, the other value is recomputed so that
the bins fill the range from Minimum to Maximum.
Figure 16-31
Selecting Bin
Width and
Number of Bins
Range
The histogram is plotted with X-Axis values from the range minimum to
maximum.
When the Automatic Range box is checked, the minimum and maximum
range values are set to the minimum and maximum values from the data
files. The range is recomputed each time the Y-Axis Parameter or display
data file is changed. cted parameter.
Figure 16-32
Selecting
Automatic Range
When the Automatic Range is not checked, you may enter the minimum
and maximum X-axis values to plot. The values you enter will take affect
when you select another edit field or press the Apply button.
The manual range values will remain the same when a different Y-Axis
Parameter or data file selected. If the histogram is blank, try switching to
Automatic Range to see the actual data range and use this information to
help in resetting the manual range.

Cumulative
When the Cumulative box is checked, a line representing the percentage of
counts less than each bin is plotted with the histogram. The cumulative line
is scaled from 0% at the bottom of the graph to 100% at the graph top.
Sub-Integer Plot
The Sub-Integer check box was designed to give a “Peak Locking” display.
Peak Locking is a bias that gives more readings near integer pixel
displacements and fewer readings near integer +0.5 locations. When the
sub-integer box is checked, the value plotted in the histogram is:
Parameter Value – round(Parameter Value)
This formula gives a value from –0.5 to +0.5. The sub-integer minimum and
maximum x-axis range is fixed to –0.5 to +0.5. When the histogram has a
high number of counts in the center (integer displacement) and few counts
at the edges (integer –0.5 and integer +0.5), peak locking may be the
reason.
The following screen shows a Sub-Integer Histogram Plot.
Figure 16-33
Sub-Integer Histogram Plot
The histogram from a single vector file may not have enough data points to
give a clear view of the pattern. To fill in the histogram, select multiple files
in the file page, and then select All on the file page before switching to the
histogram view.

Some care needs to be exercised in interpreting the peak locking graph. If
the velocity range is small (l pixel range less), the graph may look like it has
peak locking, when it is really a reflection of a small displacement range.
Increasing the time between laser pulses (dT) will increase the displacement
range and may make the measurements more accurate. A flow may happen
to have more displacements at integer pixel values than half integer values.
To test for this, change the time between pulses (dT) by a small amount
checking if the most counts are still at the integer location.
The degree of peak locking is a function of both the images and the
processing. When the particles are imaged as single pixel in size, the
processing will be peak locked because the image does not contain
enough information in the neighboring pixels to locate the particle to sub-
pixel accuracy. Displacement peak locating algorithms are typically
optimized for 3-pixel diameter particle images, so the imaging should also
be optimized to capture 3-pixel diameter particle images. In processing, the
general rule is that deformation grid processing reduces peak locking over
Nyquist, Recursive Nyquist, or Rectangular grids.
Scatter Plot
The Scatter Plot graphs the data points on two axes. Select any parameter
for the X- and Y-axis for plotting. To create a scatter plot, first open a data
file using the File page controls. Go to the Plot Type page and select the
Scatter Plot option. And the graph will switch to a scatter plot.
Figure 16-34
Selecting Scatter
Plot

Figure 16-35
Scatter Plot Screen
Figure 16-36
Changing Y-Axis
for Scatter Plot
With the Scatter page you can select the parameters for plotting on the X-
and Y-axis. Any pair of parameters can be plotted together. Seeing pattern
of the two parameters may give some insight into the flow behavior. To set
the parameter for the X- or Y-axis select it from the pull down list. The
graph range can either be set to Automatic ranging which set the limits to
display all of the data points or manual where you can enter the display
limits. To manually set the graph limits, deselect the Range and then enter
the Minimum and Maximum range value. The change takes affect when
you leave the text field by selecting a different edit field.

Troubleshooting
Included with Insight software is a full version of Tecplot® software. When
you click the “Tecplot” icon in Insight, Tecplot® software will automatically
open with a quick macro panel for easy viewing of your PIV data.
The TSI-Tecplot® macro works with several different versions of Tecplot®

software; however, the correct dll must be installed. Currently supported
Tecplot® versions are Tecplot 9, Tecplot 10, Tecplot Focus, Tecplot 360
(2006), and Tecplot 360 (2009).
If the TSI Macro does not automatically appear when you press the
“Tecplot” button in Insight software, you will need to run through the
following setup steps:
1. Open the file “Tecplot.add” in Wordpad. This file may be found in one
of the following locations, or a different location if you specified
something other than the default:
C:\Program Files\TEC90\
C:\Program Files\TEC100\
C:\Program Files\TecFocus\
C:\Program Files\Tecplot\Tec360\
2. In the Tecplot.add file, go to the bottom of the file, and type one of the
following lines. Choose the line that corresponds to the version of
Tecplot® software that you have. Then click Save. This tells Tecplot®
software to load the proper *.dll.
$!LoadAddon "TecPIV" # DLL -TSI - Insight 4G to

Tecplot9
$!LoadAddon "PIVPlot" # DLL -TSI - Insight 4G to
Tecplot10
$!LoadAddon "PIVPlot" # DLL -TSI - Insight 4G to
Tecplot 360(2006)
$!LoadAddon "FocusPIV64" # DLL -TSI - Insight 4G
to Tecplot 360(2009)
3. Now paste the *.dll file into the “bin” folder, for example:
C:\Program Files\TEC90\bin
C:\Program Files\TEC100\bin
C:\Program Files\TecFocus\bin
C:\Program Files\Tecplot\Tec360\Bin
(For Tecplot 9 it will be TecPIV.dll)

(For Tecplot 10 it will be PIVPlot.dll)
(For Tecplot 360 (2006) it will be PIVPlot.dll)
(For 32-bit Tecplot Focus it will be FocusPIV32.dll)
(For 64-bit Tecplot Focus it will be FocusPIV64.dll)For Tecplot 360
(2009) it will be FocusPIV.dll)
The dll can be found in C:\Program Files\TSI\Insight 4G\

For Further Information
For further information, contact:
Tecplot, Inc.
P.O. Box 52708
Bellevue, WA 98015
Phone: (425) 653-9393
Fax: (425) 653-9200
E-mail: [email protected]
website: http://www.tecplot.com


CHAPTER 17
Enabling Distributed
Processing (PIV and
GSV Only)
To reduce data processing time for PIV and GSV images, Insight software
allows you to perform distributed processing over a network of computers.
Insight software can be installed on another computer and run it as a
Worker PC without a security key.
To Setup your Computer for Distributed

Processing
1. In Windows® Explorer, locate the experiment folder you want to share
for distributed processing. With the default settings this would be
C:\Experiments11.
2. In the General tab enter a share name PIV_Host in this example.
Figure 17-1
Experiments11
Properties–
General Tab
17-1
3. In the Sharing tab click the Share… button.
Figure 17-2
Experiments11
Properties–
Sharing Tab

4. Select the group you want to share the PIV processing with.
Figure 17-3
File Sharing
5. Click the Share button.
6. From the Insight software main menu select Tools | Insight Setup.
7. Click Enable Distributed Processing.
8. Click the This is a Host PC button.
Enabling Distributed Processing (PIV and GSV Only) 17-3

9. In the Host Share Name: Enter the share name you entered in the
Folder Properties dialog.
Figure 17-4
Software Setup
Screen
10. Click the Test File Access button to verify the host sharing is working.

To Setup the Worker Computer for Distributed
Processing
1. On the Worker computer open Windows® Explorer browser.
2. Right click on the Network icon and select Map Network Drive.
Figure 17-5
Map Network Drive Screen
3. Click the Browse button.
4. Navigate to the Host Computer and select the PIV Host folder.
Figure 17-6
Browse for Folder
Screen

5. Click OK and Finish.
6. On the worker computer Insight software main menu select Tools |
Insight Setup.
7. Select Enable Distributed Processing.
8. Click the This is a Worker PC button.
Figure 17-7
Insight Setup Screen

9. Click the Host Experiments Directory browse button.
Figure 17-8
Browse for Folder Screen
10. Navigate to the mapped drive and click OK.
Figure 17-9
Insight Setup Screen
11. Click the Test File Access button to ensure file access.

12. The file sharing should return successfully.
13. Click the OK button.
Figure 17-10
Distributed
Processing:
Worker Mode PC
Name Screen
14. This dialog shows that the worker PC is waiting for work.
15. If you closed the Distributed Processing Worker dialog, reopen it from
the main menu by selecting Tools | Insight Setup… | OK.
16. To start distributed processing among your computers, use the Setup
Batch option in the menu bar to create a batch. When executing the
batch, distributed processing will take place.

CHAPTER 18
Developing and Using
Plugin Dynamic Link
Libraries (PIV Only)
Insight™ software comes with several algorithms to choose from for

processing PIV images. These algorithms produce good results for the
majority of experiments. However, to provide measurements in flows with
special challenges and research, user-created algorithms can be plugged
into one of these processing points.
Note: The plugins provided in Insight software were developed using

Microsoft® Visual C++ Version 6. TSI assumes that you have C++
skills to develop these plugins. The example plugins are Win32
Dynamic-Link Libraries and do not use MFC (Microsoft® Foundation
Class). Other C++ compilers should also work, but TSI has not
tested them.
Plugin Definition and Structure

A plugin is a DLL that implements a Plugin Interface. Each plugin point has
an interface that defines the methods that plugin must implement.
Standardizing the inputs and outputs of each plugin point with a DLL
interface allows Insight software to call plugin methods without having to
know anything about the algorithm used to implement the method. This
standardization allows the algorithms to be mixed and matched to optimize
processing.
The plugin structure allows you to have control of the few most critical
points in the PIV processing. The plugin author does not have to search
through large amounts of code looking for the parts that are of interest. By
isolating plugin methods, the testing and debugging of the new code is
simplified.
To create plugins quickly and easily, a set of plugin wizards has been
developed. The plugin wizards work with Visual C++ to automate the
creation of a new plugin project. The wizard-created plugins take care of
the overhead of data transfer and communication with Insight software
allowing you to concentrate on implementing your algorithm.
18-1
PluginData and Dynamic Link Libraries
(DLL) Plugin Parameters
Features of the plugins include PluginParameters and PluginData.
Plugin Parameters
Plugin parameters are used to control the processing algorithm. A plugin
parameter may be of type boolean, integer, floating point, or string. The
DefaultPeakEngine Correlation Engine parameter is an example of a string
Plugin Parameter, the Rectangular grid uses a boolean Plugin parameter
to control if recursive processing is to be used, integer Plugin parameters
set the first pass spot offsets, and a float Plugin parameter sets the spot
size reduction between passes. With plugin parameters you can develop
one plugin that can be optimized depending on the circumstances.
Plugin Data
The Plugin data feature is used to make user-defined measurements
during processing. Data is created for each vector. This data can be used
as a signal-to-noise ratio for validation, exported to the vector file, and
viewed in the process monitor. The plugin data class holds the following
information: measured value, and if it passed or failed the SNR test, data
name, plugin name, export on, SNR filter, SNR threshold, SNR greater
than or less than test.
SNR Vector Validation

SNR vector validation is used to eliminate vectors that have poor signal-to-
noise ratio during processing. Filtering out bad vectors during processing
by measuring input spot or correlation values produces a vector field with
less need for validation based on the velocity. SNR validation can measure
and use any parameter. Two common SNR tests are Minimum Average
Spot Intensity and the Correlation Displacement Peak Height to highest
noise-to- peak- ratio. The average spot intensity is measured in the Base
Spot Mask. The Peak To Noise Ratio is measured in the Base Peak Mask.
Other SNR measurements may be added by plugins.

Plugin Development Kit
Insight software during installation installs the files you need to create your
own plugins and example plugin code in the TSI\Insight 4G\PluginSDK
folder. The following lists the contents of this folder:
Folder Contents
BaseEngine Contains base classes for deriving your plugin.
Interface Contains interfaces that are used in exchanging data
between Insight software and the plugin.
ExampleCode Contains example source code showing an example
implementation for each plugin point. ExampleCode
DLLs can be used as the starting point for developing
your own plugin.
Installing TSI Plugin Wizards into

Visual C++
The following plugin wizards are installed in the Insight
4G\PluginSDK\Wizard folder:
 TSI Grid Wizard.awx
 TSI Spot Mask Wizard.awx
 TSI Correlation Wizard.awx
 TSI Peak Wizard.awx
To use the wizards for Visual C++, copy them from the PluginSDK\Wizard
folder into your Visual C++ Template folder default location:
C:\Program Files\Microsoft Visual

Studio\Common\MSDev98\Template.
Plugin Coordinates and Units

The plugin processing is done in pixel units. The world units of m/s and mm
are not used in the plugins. Velocity measurement in m/s are made by
computing all of the particle image displacements and vector locations in
pixel units, and then converting to mm location and m/s velocity units as
the vector file is saved to disk.
The plugin image, spot and correlation map coordinates have the (0, 0)
pixel in the lower left corner and (Width, Height) in the upper right corner.
This is different from the standard image coordinate system with the (0, 0)
pixel in the upper left corner and the (Width, Height) pixel in the lower right
corner. Using the Y increasing up coordinates better matches standard
world coordinates and simplifies the conversion from pixels to mm and m/s.
Developing and Using Plugin Dynamic Link Libraries (PIV Only) 18-3
TSI Plugin Wizard Generated Code
The code that the TSI Plugin Wizards generate have two sections—DLL
export methods that communicate with Insight software and a C++ Class
where the real work of the plugin is done. The DLL export methods simply
call a class method with the same parameter list, with the exception of
InterfaceImplemented() that returns the interface version number. This
structure gives you all of the advantages of programming in C++ without
having to program the overhead of DLL communication.
When creating a plugin, first derive and compile the plugin using the wizard
generated code to verify that the plugin is functioning and connected to
Insight software. Then edit or delete the wizard methods and data as you
implement your algorithm.
The classes are named CustomGridClass, CustomSpotMaskClass,

CustomCorrelationClass, and CustomPeakClass.
Output Folder
C:\Program Files
\TSI Insight 4G
\Processor
Plugin Bass Class Class \Kernel\Plugin
Grid BaseGrid CustomGridClass GridEngine
SpotMask BaseSpotMask CustomSpotMask SpotMask-Engine
Correlation BaseCorrelator Custom- Correlation-Engine
CorrelationClass
Peak BasePeak CustomPeakClass PeakEngine
ExampleCode
Each ExampleCode subfolder contains a plugin implementation. The folder
includes the plugin CPP and H source code files. StdAfx.cpp and StdAfx.h
files are used by Visual C++ for precompiled headers. A Visual C++ project
workspace *.DSW and a Visual C++ project *.DSP are also included.

Plugin Description
GaussianMask Applies a Gaussian filter to the input Spot A reducing the
pixel intensity away from the center of the spot. The
GaussianMask decreases the relative strength of spot edge
particles. The correlation of edge particles where different
amounts of the particle are included in Spot A and Spot B
increases the correlation peak location error. GaussianMask
demonstrates the use of PluginData with a minimum average
spot intensity measurement.
RectangularGrid An example that shows the use of float, integer and boolean
PluginParameters. It has float PluginParameters to set the
grid spacing based on the width and height of Spot A. It has a
boolean PluginParameter to turn recursive processing on or
off. It has a float PluginParameter to set the spot size
reduction between processing passes. It has a boolean
PluginParameter to set central-difference or forward-
difference offset. And it has integer PluginParameters to set
the first processing pass offset distance.
DirectCorrelator Implements the direct correlation. This is the code that the
TSI correlation wizard creates without any modifications.
GaussianPeak Locates the displacement peak by doing a 3-point Gaussian
fit in the X and Y directions around the highest correlation
map pixel. This the TSI Peak Wizard code without any
modifications.
Creating a Plugin Using a Wizard

You can start a plugin project with Microsoft® Visual C++® 6 using one of
the TSI-provided wizards.
A plugin wizard has been created for each plugin point. The plugin wizard
generates a project that is linked to the PluginSDK folder so that it can
simply be built and run with Insight software. Each of the wizards
implements an algorithm so that the default wizard code is a valid plugin.
The generated code can then be modified to implement your algorithm.
The creation of a plugin for any of the plugin points is the same, with the
exception of choosing the TSI Wizard in the New Project dialog, and
setting the output file path to the correct engine folder in the project
settings dialog. The following steps illustrate how to create the plugin
project using the Gaussian Spot Mask Plugin, as an example. The code is
included in your C:\Program
Files\TSI\Insight 4G\PluginSDK\ExampleCode\GaussianMask folder.
1. From the Visual C++ main menu select File | New | Projects and
select TSI Spot Mask Wizard. The other wizards included are: TSI
Correlation Wizard, TSI Grid Wizard or TSI Peak Wizard.
Figure 18-1
Using TSI Spot Mask Wizard
2. Enter the name in the Project name box and location in the Location
box.
3. Click OK. If you do not see the TSI Plugin Wizards, see “Installing TSI
Plugin Wizards into Visual C++” section of this chapter. The New
Project Information dialog appears describing the plugin being created.

Figure 18-2
New Project Information Dialog
4. Click OK.
If Insight software was installed in the default folder you are prompted
to specify if the plugin should be created in that folder. To create the
plugin in the default folder, click Yes.
If Insight software was not installed in the default folder or to select a
different folder to create the plugin in, navigate to the folder where the
executable file for Insight software and the PluginSDK folder are
installed.
After the project files are created, perform the following step.
5. From the Visual C++ main menu, select Project | Settings | Debug.
The following dialog box appears.
C:\Program Files\TSI\Insight 4G\Insight4G.exe
Figure 18-3
Debug Tab – Project Settings Screen
6. Make the following selections:
Field Selection
Settings For All Configurations.
Executable for Enter the full path to Insight 4G.exe
debug session:
7. Click General tab and in the Output files box, specify the path to
the plugin engine folder. In our example, C:\Program Files\Insight
4G\TSI\SpotMaskEngine. The default plugin engine folders are:
C:\Program Files\TSI\Insight 4G\GridEngine

C:\Program Files\TSI\Insight 4G\SpotMaskEngine
C:\Program Files\TSI\Insight 4G \CorrelationEngine
C:\Program Files\TSI\Insight 4G \PeakEngine

C:\Program Files\TSI\Insight 4G\SpotMaskEngine
Figure 18-4
General Tab – Project Settings Screen
8. Click OK. The plugin DLL compiles and builds. If there are errors,
verify that the output DLL file is in the correct folder and the
PluginSDK files are being located by doing the following:
Click the Visual C++ Execute Program <Ctrl>+<F5> button or the
Go <F5> button to start Insight software. If Insight software does
not start, verify the Project | Settings | Debug | Executable for
debug session value.
Test
Figure 18-5
PIV Processor Setup
In Insight software, select the PIV Tab and choose

Processor Setup | Processor Settings, | Spot Mask Engine
drop down list. The new Spot Mask should be listed, if not, verify
that the path: Visual C++ Project | Settings | General | Output
values is C:\Program Files\TSI\Insight
4G\Processor\Kernel\Plugin\SpotMaskEngine.
9. Select the new plugin and process an image to verify that the
plugin is working.
You are now ready to start customizing the files to implement your
own algorithm.
To debug or trace through the code, set a break point at the line of
code where you want to start debugging. Then with Visual C++ set
for Win32 Debug launch Insight software using the Go <F5>
button. When the line of code with the break point is executed,
Insight software pauses as you trace through the code and use the
debugger. When you click the Go <F5> button again, Insight
software continues processing until the next break point.

Customizing the Gaussian Spot Mask Engine
Two Plugin Parameters are used in this plugin; a float parameter that
controls the Gaussian mask radius, and a bool parameter that turns mean
subtraction on or off.
To Customize the Gaussian Spot Mask

1. Open the GaussianMask.h file.
Figure 18-6
GaussianMask.h
2. Declare two static variables of type IPluginParameter*

s_pSubMeanParam and s_pGaussRadParam, and one method
ApplyGaussianMask() inside the CustomSpotMaskClass.
3. Open the GaussianMask.cpp file.
Figure 18-7
GaussianMask.cpp
4. Add #include <math.h> at the top of the file. It is required for the
exponential (exp) function.
5. Define the static members s_pSubMeanParam and
s_pGaussRadParam.

Figure 18-8
GaussianMask.cpp Plugin Settings
6. In the CustomSpotMask::InitalizePluginSettings() method, initialize the

two IPluginParameter* variables s_pSubMeanParam and
s_pGaussRadParam and set the default values.
7. In the GaussianMask.cpp file, edit the ApplySpotMask() method.
Figure 18-9
GaussianMask.cpp Apply Spot Mask
8. Create two float variables aSub and bSub and initialize to 0. Check the
s_pSubMeanParam value and if true, set aSub and bSub to the
average spot intensities.
9. Change the SpotA CopySpot() method to ApplyGaussianMask(), and
change the SpotB CopySpot() method to SubtractCopySpot(). The
ApplyGaussianMask method will be entered in the next step, the
SubtractCopySpot subtracts the input value, bSub, from each pixel.
10. Type in the CustomSpotMaskClass::ApplyGaussianMask() method.

Figure 18-10
GaussianMask.cpp Apply Gaussian Mask
This method first checks that a valid Gaussian radius was entered and if
not uses SubtractCopySpot(). The Gaussian equation parameters are then
computed and each pixel value is multiplied by a Gaussian weighting
factor.
To debug your plugin, set break point and then run Insight software from
Visual C++ in debug mode by launching Insight software with the Go <F5>
button.
Plugin SDK Files and Classes
The following lists the files in the PluginSDK folder and their classes:
BaseEngine Folder
The BaseEngine folder contains the following plugins:
BasePlugin
This is the base class for all plugins. It implements the following methods
common to all plugins.
Insight DLL Interface Methods

LPCSTR GetErrorMessage(int handle); Returns the error message string.
int InterfaceImplemented(int pluginID); Returns true if the DLL implements the specified
pluginID.
int CreatePlugin(); Returns the plugin’s handle 0 if unsuccessful.
void
CreatePluginSettings(IPluginSettingFactory2*
pFactory);
bool DestroyPlugin(int handle); Destroys a previously created plugin using the
plugin’s handle. Returns true if successful.
Useful Plugin Methods
bool SetPluginData(LPCSTR pluginName, LPCSTR Finds the measured plugin name and data name to
dataName, float value, IPluginData** pDataArray, the PluginData array, sets and validates the value.
bool& pass) SNR pass or fail is returned in pass. Returns false if it
could not find a match between input names to the
PluginData Array. (Deprecated)
bool SetPluginData2(int pluginID, LPCSTR Finds the measured pluginID and data name to the
dataName, float value, IPluginData2** PluginData array, sets and validates the value. SNR
pDataArray, bool& pass) pass or fail is returned in pass. Returns false if it
could not find a match between input names to the
PluginData Array.
LPCSTR GetErrorMessage(void) Returns the error message.

BaseGrid
This is the base class for GridEngines, derived from the BasePlugin. It
implements the following common grid methods and computations:

bool SetGridParameters( Stores the processing setup parameters.
int handle,
const int aStartSpotWidth,
const int aStartSpotHeight,
const int bStartSpotWidth,
const int bStartSpotHeight,
const int aFinalSpotWidth,
const int aFinalSpotHeight,
const int bFinalSpotWidth,
const int bFinalSpotHeight,
const RECT aoi,
const int imageWidth, const int imageHeight);
bool GetGridSize(int handle, int& vecColumns, Computes and returns the grid size for the current
int& vecRows, int& spotColumns, int& spotRows, pass along with the spot columns, spot rows and
bool& isFinalPass); whether this is the final pass or not.
bool FillInGrids(int handle, IProcessGrid* Fills the grids using the passed in grid mask.
pProcessGrid, IPluginVectorField2* pVF, (Deprecated)
IPluginVectorField2* PreviousPassVF,
IPluginProcessingMask* GridMask)
bool FillInGrids2(int handle, IProcessGrid* Fills the grids using the passed in grid mask.
pProcessGrid, IPluginVectorField2* pVF,
IPluginVectorField2* PreviousPassVF,
IPluginProcessingMask* pGridMask, bool
invertedGridMask)
bool GetIsMultipass(int handle) Returns true if grid engine is multipass.
bool GridSizeUniformSpacing( Calculates the number of vectors that fit with the input
const int xSpacing, const int ySpacing, parameters.
const int aSpotWidth, const int aSpotHeight,
const int bSpotWidth, const int bSpotHeight,
const RECT aoi, int& columns, int& rows);
Bool FillUniformSpacingGrid( Sets the Spot A and Spot B dimensions and locations
IprocessGrid* pProcessGrid, and stores in the Process Grid using the Previous
IpluginVectorField* pVF, Pass Vector field to offset the spots.
IpluginVectorField* pPreviousPassVF, Initializes the vector field.
IpluginProcessingMask* pGridMask,
const int xSpacing, const int ySpacing,
const int aSpotW, const int aSpotH,
const int bSpotW, const int bSpotH,
const RECT aoi, const int imageWidth,
const int imageHeight,
bool isCentralDifferenceOffset);
RECT GetVectorArea(const int spotWidth, const Calculates the image range for vectors. The vectors
int SpotHeight, const RECT aoi); cannot be closer than ½ spot from the edge of the
area of interest.
Void ComputeOffsets(float xVec, float yVec, Calculates the spot offsets base on a vector from the
IgridPoint* pGridPt, int& aXOffset, int& Previous Pass Vector field.
aYOffset, int& bXOffset, int& bYOffset, int&
offsetChoiceCode);
Void ReduceSpotSize(float factor, bool& Calculates the spot sizes for the next processing
isFinalSize); pass. If the spots are at the final size returns
isFinalSize = true.
BaseSpotMask
The Base class for SpotMaskEngines is derived from BasePlugin. It
implements the following common SpotMask methods:
Insight DLL Interface Methods Description

bool MaskSpots(int handle, ISpotImage* pSpotA, Applies masking to the input spots. (Deprecated)
ISpotImage* pSpotB, ISpotImage* pMaskedSpotA,
ISpotImage* pMaskedSpotB, IPluginData2**
pDataArray, bool& pass)
bool MaskSpots2(int handle, ISpotImage2* pSpotA, Applies masking to the input spots.
ISpotImage2* pSpotB, ISpotImage2* pMaskedSpotA,
ISpotImage2* pMaskedSpotB, IPluginData2**
pDataArray, bool& pass)
Bool CopySpot(ISpotImage* srcSpot, Copies the pixel values from the source spot the
ISpotImage* destSpot); destination spot. Returns false if the spots are
different sizes.
float GetAvgIntensity(ISpotImage* pSpot); Returns the average pixel intensity for the spot.

BaseCorrelator
The Base class for CorrelationEngines, derived from BasePlugin.
Implements the following common correlation methods.

bool Correlate(int handle, ISpotImage* pSpotA, Correlates input spots into correlation map.
ISpotImage* pSpotB, ICorrelationMap2*
pCorrelationMap, const int dxMax, const int
dyMax, float& xZero, float& yZero, bool& pass)
bool GetMapSize(int handle, ISpotImage* pSpotA,
ISpotImage* pSpotB, int& dxMax, int& dyMax, int&
mapWidth, int& mapHeight)
bool ComputeMinimumMapSize( Calculates the smallest correlation map that will hold
ISpotImage* pSpotA, ISpotImage* pSpotB, the range of displacements in the correlation. Returns
int& dxMax, int& dyMax, the map size by reference. Returns false if there is an
error computing the map size.
int & mapWidth, int & mapHeight);
BasePeak
The Base class for PeakEngines, derived from BasePlugin. It implements
common peak search methods.

bool LocatePeak(int handle, ICorrelationMap2* (Deprecated)
pCorrelationMap, IPluginData2** pDataArray,
const float xZero, const float yZero, const int
dxMax, const int dyMax, float& dx, float& dy,
bool& pass)
bool LocatePeak2(int handle,
ICorrelationMap2* pCorrelationMap,
ICorrelationMap2* pAutoCorrelationMapA,
ICorrelationMap2* pAutoCorrelationMapB,
IPluginData2** pDataArray,
const float xZero, const float yZero, const int
dxMax, const int dyMax, float& dx, float& dy,
bool& pass)
Bool FindPeakPixel(ICorrelationMap2* pMap, Finds the highest intensity correlation map pixel in the
const float xZero, const float yZero, search area. If no peaks found x and y are –1.
const int dxMax, const int dyMax, Returns false if error.
int& x, int& y);
Bool GetPeakNeighborhood5( Finds the highest intensity pixel in the correlation map
ICorrelationMap2* pMap, search area and returns it and the four nearest
const float xZero, const float yZero, neighbors, the pixel used for 3-point curve fitting. If no
peak was found x and y = -1. Returns false if error.
const int dxMax, const int dyMax,
float& vPeak, float& vLeft, float& vRight,
float& vUp, float &vDown, int& x, int& y);
Void SetOutOfBoundsErrorMsg(int x, int y, A common error in peak search is looking outside the
LPCSTR methodName); bound of the correlation map. This method is used to
generate a formatted error showing the out-of-bounds
location you tried to access.
Float MeasurePeakToNoiseSNR(void); Computes the peak pixel intensity divided by the
highest noise peak pixel intensity. A common SNR
test for filtering out bad vectors. Returns the ratio.
void SetSearchArea(ICorrelationMap2* pMap, Use to calculate the range of correlation map pixel to
const float xZero, const float yZero, search for the peak.
const int dxMax, const int dyMax);
bool FindPeaks(ICorrelationMap2* pMap); Scans the image and finds the local peaks, storing
the parameters of the four largest peaks.
Returns false for error.
bool IsLocalPeak(ICorrelationMap2* pMap, Returns true if the pixel is higher than all of its
int x, int y, float v, float& sum); neighbors.
Sum is the sum of the pixel and its neighbors, use as
a tie-breaker if two peaks have the same height.
void StorePeak(int x, int y, float peakValue, Store the peak location in the peaks array.
float sum);
void ClearPeakArray(); Resets the peak array
void ShuffleDown(int i); Use to sort peaks in the array.
class PeakRecord Use for the peak array.
Interface Classes
The following interface classes and class methods are described. For
detailed information, including comments on the use of these methods,
refer to the *.H files in the Insight software \PluginSDK\Interface folder.
 ICorrelationMap2
 IGridPoint
 IPluginData2
 IPluginParameter
 IPluginSettingFactory2
 IPluginVector2
 IPluginVectorField2
 IProcessGrid
 IProcessingMask
 ISpotImage
 PluginID
The following shows an example of a *.H file view with Visual C++.

Figure 18-11
Interface IGridPoint.h
Class Description
ICorrelationMap2 The correlation map is the output from the
CorrelationEngine and the input to the
PeakEngine. The correlation map is a 2-D-array
of floating point pixels. The particle image
displacement is distance from the zero pixel to
the particle image displacement peak plus the
offset.
Class Description
ICorrelationMap2 Methods
void SetSize(int width, int height);
bool SetPixelValue(int x, int y, float value);
bool AddToPixelValue(int x, int y, float value);
bool GetPixelValue(int x, int y, float& value);
int GetWidth(void);
int GetHeight(void);
void SetZeroPixel(const float xZero, const float yZero);
bool GetZeroPixel(float& xZero, float& yZero);
void SetMaxDisplacement(const int dxMax, const int dyMax);
bool GetMaxDisplacement(int& dxMax, int& dyMax);
void SetSpotOffset(const int xOffset, const int yOffset);
bool GetSpotOffset(int& xOffset, int& yOffset);
ICorrelationMap2& operator +=(ICorrelationMap2& correlationMap);
IGridPoint A grid point is a definition of a pair of spots to be

processed. Each spot has a width, height and x,
y location. Translating the spots x, y locations
sets offsets. The choice code is used to show if
a grid point should be processed. CheckBound
is used to find out if one of the spots has been
offset beyond the image bounds and its choice
code should be set to CC_BOUNDS.
GridPointMethods
.Bool SetSpot(int index, int left, int bottom, int width, int height);
bool GetSpot(int index, int& left, int& bottom, int& width, int& height);
bool CheckBounds(const int imageWidth, const int imageHeight;
void SetChoiceCode(const int code);
int GetChoiceCode(void);
int GetXOffset(void);
int GetYOffset(void);
IPluginData2 A measured value from one of the plugins. Stored in

the IpluginVector.

Class Description
IPluginData2 Methods
Void SetValue(float value);
float GetValue(void);
void SetExport(bool writeToFile);
bool GetExport(void);
void SetSnrFilter(bool snrValidate);
bool GetSnrFilter(void);
void SetSnrThreshold(float thresholdValue);
float GetSnrThreshold(void);
void SetSnrGreaterOrLessThan(bool isGreaterThan);
bool GetIsGreaterOrLessThan(void);
void SetDataName(LPCSTR dataName);
LPSTR GetDataName(void);
void SetPluginName(LPCSTR pluginName);
LPSTR GetPluginName(void);
bool Validate(void);
bool ValidateValue(float value);
int GetPluginID(void);
IPluginData2& operator =(IPluginData& pluginData);
IPluginData2& operator +=(IPluginData& pluginData);
IPluginParameter PluginParameters are used to give you control over

plugin settings. PluginParameters may be Boolean,
float, integer or string variables.
IPluginParameter Methods
enum ParameterType
{
PLUGIN_PARAMETER_BOOL,
PLUGIN_PARAMETER_FLOAT,
PLUGIN_PARAMETER_INT,
PLUGIN_PARAMETER_STRING
};
LPCSTR GetParameterName(void);
LPCSTR GetPluginName(void);
int GetParameterType(void);
bool SetBoolValue(bool b);
bool GetBoolValue(void);
bool SetFloatValue(float f);
float GetFloatValue(void);
bool SetIntValue(int i);
int GetIntValue(void);
bool SetStringValue(LPCSTR string);
LPCSTR GetStringValue(void);
Class Description
IPluginSettingFactory2 Used by a plugin to create IPluginDataConfig
and IPluginDataConfiguration objects.
IPluginSettingFactory2 Method
void CreatePluginDataConfig(
int pluginID,
LPCSTR pluginName,LPCSTR dataName,
bool isSnrFilter, bool isGreaterThan,
float snrThreshold, bool exportToFile);
IPluginParameter* CreatePluginParameter(
int pluginID,
LPCSTR pluginName,
LPCSTR ParameterName,
int parameterType);
IPluginDataConfigurationArray* GetPluginDataConfigArray();
IPluginVector2 The vector during processing. The

IPluginVector contains the location, velocity,
choice code, ICorrelation Map and
IPluginData. The ICorrelationMap is only
available during processing until the next
image is loaded. The IPluginData2 is also only
available during processing until the next
image is loaded if Export is false.
IPluginVector2 Method
typedef enum
{
CC_BOUNDS = -4,
CC_NOT_SET,
CC_REMOVED,
CC_SNR,
CC_BLANK,
CC_1,
CC_2,
CC_3,
CC_INTERPOLATED,
CC_SMOOTHED,
CC_MAX
} ChoiceCode;
int getChoiceCode();
LPCSTR getChoiceCodeString(void);
float getU();
Float getV();
float getX();
float getY();

Class Description
IPluginVector2 Method (continued)
IPluginVector& operator = (IPluginVector& vector);

void setChoiceCode(int choiceCode);
void setU(float u);
void setV(float v);
bool isHole();
void setX(float x);
void setY(float y);
int getPluginDataArraySize(void);
bool getPluginDataArray(
IPluginData** pDataArray,
int capacity);
void setPluginData(LPCSTR pluginName,
LPCSTR dataName, float value);
ICorrelationMap* getCorrelationMap(void);
void setSpotOffset(const int xOffset,
const int yOffset);
bool getSpotOffset(int& xOffset,
int & yOffset);
ICorrelationMap2* getCorrelationMap2();
bool getPluginDataArray2(IPluginData2**
pDataArray, int capacity);
ICorrelationMap2* getAutoCorrelationMapA();
ICorrelationMap2* getAutoCorrelationMapB();
IPluginVectorField2 2-D array of IPluginVector objects.
Class Description
IPluginVectorField2 Methods
int getGridColumns();
int getGridRows();
int getImageHeight();
int getImageWidth();
void setImageDimensions(int width,
int height);
IPluginVector& getNearest(
float imageX, float imageY);
IPluginVector& get(int gridX, int gridY);
IPluginVectorField& operator =
(IPluginVectorField& vectorField);
void set(int gridX, int gridY,
IPluginVector& vector);
void write(const char* fileName) throw();
IPluginVector2& getVector2(int gridX,
int gridY);
IPluginVectorField2& operator =
(IPluginVectorField2& vectorField);
void setVector2(int gridX, int gridY,
IPluginVector2& vector);
IPluginVector2& getNearestVector2(float imageX, float imageY);
IProcessGrid 2-D array of IGridPoints.

IProcessGrid Methods
bool SetGridDimensions(int columns,
int rows);
bool GetGridDimensions(int& columns,
int& rows);
IGridPoint* GetGridPoint(int column,
int row);
int GetNumberGridPoints(void);
IGridPoint* GetGridPoint(int index);
IProcessingMask The processing mask is created using the ROI tab

and specified in the Processor Setup dialog. Points
with masked pixels are removed or included from the
vector field.
IProcessingMask Methods
bool getContainsMaskedPixel(int x, int y);
bool getContainsMaskedPixel(int left, int bottom, int width, int height);

ISpotImage2 The pixel values of one of the spots being processed.
ISpotImage is used for both the raw image spots and
the masked spots. The pixel are stored as int with a
range of
–216 to 216 for all image bit depths.

Class Description
SpotImage Methods
void SetSize(int width, int height,
int bitsPerPixel);
bool SetPixelValue(int x, int y, int value);
bool GetPixelValue(int x, int y, int& value);
int GetWidth(void);
int GetHeight(void);
int GetBitsPerPixel(void);
int GetGrayScaleMax(void);
void ZeroPixels();
void SetXSpotMaskOffset(float xShift);
void SetYSpotMaskOffset(float yShift);
float GetXSpotMaskOffset();
float GetYSpotMaskOffset();
PluginID Enumerates the ID numbers for the plugin points.
PluginID Methods
#define ENGINE_GRID 0x0001
#define ENGINE_SPOT_MASK 0x0002
#define ENGINE_CORRELATION 0x0004
#define ENGINE_PEAK 0x0008
Data Files
The following lists the different types of data files and settings stored in
Insight software.
File Description
user_defined_name.pivproc Saves PIV processor settings in these files.
user_defined_name.plifproc Saves PLIF and spray processor settings in these
files.
user_defined_name.pivcond Saves PIV image conditioning settings in these
files.
user_defined_name.plifcond Saves PLIF image conditioning settings in these
files.
user_defined_name.pivval Saves PIV validation settings in these files.

APPENDIX A
I n s i g h t Data Files
This appendix describes the file naming convention and lists all the data
files generated in Insight software.
Insight Software F i l e n a m i n g
Insight™ software uses a filenaming system to help organize the many files
generated in an experiment. The filenaming system allows many filenames
to be derived from the image filenames, so that you do not have to enter
filenames for each process. A typical full filename for an image is listed in
Table A-1 below.
Table A-1
Filename Components
C:\Experiments7 Full filename
\Demo000000.T001.D001.P001.H001.La.ti
f
C:\Experiments7 Directory where the files are stored. This is the name specified when Insight
software is setup. In this example, Experiments 7 is the directory where
these image files are stored.
Cap000000.T001.D001.P001.H001.Ra.tif Filename
Filename Components
Cap RunName
000000 Sequence Number
All images that are processed to create a vector field have the same frame
number.
T0001 Traverse.
D0001 Delay
P001 Process
H001 Hardware configuration
R Camera Identifier used for two camera systems.
A camera ID is added to the filename if two-camera image capture is being
used. L is used for the left camera and R is used for the right camera. If you
have a two-camera system but are capturing a sequence using only the left
or only the right camera, no Camera ID character is used. If you have a
single camera system, no Camera ID character is added to the filename.
A-1
Table A-1
Filename Components
a Frame Straddle Frame Identifier.
If Frame Straddle Image capture mode is selected, a Frame ID character is

added to the filename. The character A designates the first frame and B
designates the second frame of a frame straddle image pair. If Frame
Straddle mode is not selected, no Frame ID character is used in the
filename.
.tif File extension
Table A-2
Number of Images with Same Sequence Number for Image
Capture Modes
Number of Images
Number Frame Straddle with same Sequence
Cameras Capture Number
1 no 1
1 yes 2
2 no 2
2 yes 4
Note: Do Not End Family Name with a Number

The sequence number is determined by reading the filename backwards. It
does not assume a fixed number of digits for the sequence number. If the
family name ends with a digit, that character is interpreted as part of the
sequence number, not the family name. This may not be desired.
Insight Software F i l e s
The following table lists all the files that are generated and used by Insight
software. Some of these files are described in detail in this appendix.
File Extension Description

Tif Standard image file.
Mac A macro, listing a number of commands or

instructions programmed by you that can be
executed with a single mouse click. See Chapter
11, “Programming Macros” for details.
Vec 2-D Vectors: a set of vectors produced by PIV
processing. It may also include summarized PLIF
data extracted from a PLIF processed image.
See “Vector (*.VEC) Files” in this appendix.
Par Particles: a set of particles produced by SUPER-
RESOLUTION PARTICLE VELOCIMETRY or
GSV processing.
A-2 Insight 4G Global Image Acquisition, Analysis and Display Software

File Extension Description
V3d 3-D Vectors: a set of 3-D vectors produced by
PIV processing. See “3-D Vector Files.”
Exp Notes and settings of an Experiment.
Run Notes and settings of a Run.
DIPAMask, GSVMask, Processing Mask. A set of rectangle, line or
PXVMask, PLIFMask polygon region of interest masks. They can be
saved and re-used.
DIPA2DCalib, GSV2DCalib, 2-D Spatial Calibration File.
PLIFCalib,
PXV2DCalib
PLIFCalib, PXV3DCalib 3-D Perspective Calibration File.
DIPAPre, GSVPre, PLIFPre, Preprocessor Pipeline Configuration File.
PXVPre, SAUTOMAPPre
DIPAKernel, GSVKernel, Kernel Process Configuration File.
PLIFKernel, PTVKernel,
PXVKernel,
SAUTOMAPKernel
PLIFPost, PXVPost, Post Processor Pipeline Configuration File.
SPIVPost
Capture Capture Settings File. All timings and settings
required to execute a capture are in this file.
Batch A batch, listing a number of macros, program by
you, that can be executed and which may be
across different experiments.
The following describes the format and contents of some of the files listed
in the above table.
Vector (*.VEC) Files

The vector file is created by Insight software when an image is processed,
or when vector fields are averaged together. The vector files may be in
units of mm and m/s, or in pixels depending on the calibration mode
selected. The file header line labels the data columns and shows which
measurement units were used. In addition to the vector location, a vector
choice code value is recorded in the vector record. This choice code shows
where the vector came from and if it should be considered a good vector. If
the vector should not be used, the velocity value is stored as 9.9e9. This
very large velocity value was chosen to create obvious errors should you
mistakenly include it in any calculations.
Insight Data Files A-3

Choice (CHC) Codes
The choice code in the velocity record identifies some information about
the vector. Only vectors with positive CHC codes should be accepted in
velocity field computations. The number of vectors with each choice code
can be found in Insight software by using the PIV Validation | Statistics
dialog, or the Statistics List File. The following lists the choice codes.
Table A-3
Choice Codes
CHC Description
1 Highest correlation peak used for vector.
2 2nd highest correlation peak used for vector.
3 3rd highest correlation peak used for vector.
4 Interpolated vector.
5 Smoothed Measured Vector, was a code 1, 2 or 3 before smoothing.
0 Temporally Blank. Vector did not pass a validation criteria and is waiting
to be filled with an interpolated vector.
-1 SNR Fail. The vector was removed because the correlation did not pass
the SNR validation criteria defined in the Process | Setup | Signal Levels
dialog.
-2 Removed point. The point has been disabled by the Process |
Polynomial Grid, or other removed point from grid.
-3 Bounds, 3-D Vector outside 2-D vector field overlap range. *.V3D file, or
the 2-D vector is out of the image bounds after spot offset.
-4 Vector has not been set.
Error Vectors
Velocity values of 9.9e9, indicate an error velocity. This is not a valid
velocity value. Following shows an error vector record:
5.200000, 49.600002, 9999899648.000000, 9999899648.000000, -1
X Position mm 5.200000
Y Position mm 49.600002
U Velocity m/s 9999899648.000000 (9.9e9)
V Velocity m/s 9999899648.000000 (9.9e9)
Choice Code -1 Non-positive choice code indicates error vector.
Pixel *.VEC Files

A pixel displacement *.VEC vector file is created by Insight software when
an image is processed in the Pixel Measurement mode. In this mode, the
vector locations are in pixels, with the lower left image corner being pixel
(0, 0) and the upper right corner being the image size in pixels (width,
height). The U and V velocities are pixel displacements. The dT time
between laser pulses is not used in the Pixel*VEC file.

A graph of the pixel VEC field looks the same as velocity VEC graphics,
except for the axis labels in Tecplot® software. Using pixel displacements is
recommend during experiment setup when you are changing the image
field of view and finding the optimum dT. Once the experiment is
configured, you should measure the field of view and make velocity
measurements.
For Stereoscopic PIV, the left and right images must be processed in pixel
units. The vector field combination converts from pixel to velocity.
Pixel *.VEC Header Line

The following shows the header line information.
TITLE="\D:\Experiments\Deg\Vector\Deg00000L.vec" Filename
VARIABLES= "X pixel", "Y pixel", "U pixel", "V pixel", Data Column names. Note that the locations and
"CHC" displacements are in pixels.
ZONE Zone
DATASETAUXDATA Application="PIV" Application type, Image size, Length Unit, Origin and
DATASETAUXDATA SourceImageWidth="128" Time Unit.
DATASETAUXDATA SourceImageHeight="128"
DATASETAUXDATA LengthUnit="pixel"
DATASETAUXDATA OriginInImageX="0.000000"
DATASETAUXDATA OriginInImageY="127.000000"
DATASETAUXDATA TimeUnit="deltaT"
DATASETAUXDATA SecondaryPeakNumber="0"
DATASETAUXDATA DewarpedImageSource="0"
I=5, J=5, Number of data columns and rows
F=POINT Data record format
Pixel *.VEC File Listing

The following lists the information in the vector file.
TITLE="C:\Experiments7\Experiment PIV
Demo\Sim\Analysis\sim000001.T000.D000.P001.H001.L.vec" VARIABLES="X pixel", "Y pixel", "U
pixel", "V pixel", "CHC", DATASETAUXDATA Application="PIV" DATASETAUXDATA
SourceImageWidth="128" DATASETAUXDATA SourceImageHeight="128" DATASETAUXDATA
LengthUnit="pixel" DATASETAUXDATA OriginInImageX="0.000000" DATASETAUXDATA
OriginInImageY="127.000000" DATASETAUXDATA TimeUnit="deltaT" DATASETAUXDATA
SecondaryPeakNumber="0" DATASETAUXDATA DewarpedImageSource="0" ZONE I=7, J=7, F=POINT
16.000000, 112.000000, -0.480119, -0.782582, 1
32.000000, 112.000000, -0.504793, -0.438491, 1
48.000000, 112.000000, -0.475778, -0.284266, 1
64.000000, 112.000000, -0.438533, 0.108856, 1
80.000000, 112.000000, -0.508797, 0.381481, 1
96.000000, 112.000000, -0.444264, 0.694730, 1
112.000000, 112.000000, -0.453020, 0.938349, 1
16.000000, 96.000000, -0.132720, -0.712976, 1
32.000000, 96.000000, -0.119912, -0.462049, 1
48.000000, 96.000000, -0.204022, -0.223730, 1
64.000000, 96.000000, -0.219707, 0.122304, 1

80.000000, 96.000000, -0.249393, 0.385273, 1
96.000000, 96.000000, -0.210901, 0.623437, 1
112.000000, 96.000000, -0.187510, 0.920299, 1
16.000000, 80.000000, 0.093866, -0.722569, 1
32.000000, 80.000000, 0.176764, -0.454197, 1
48.000000, 80.000000, 0.088524, -0.211089, 1
64.000000, 80.000000, 0.113216, 0.179897, 1
80.000000, 80.000000, 0.141416, 0.451242, 1
96.000000, 80.000000, 0.017143, 0.703499, 1
112.000000, 80.000000, 0.095544, 0.979569, 1
16.000000, 64.000000, 0.360349, -0.719723, 1
32.000000, 64.000000, 0.394953, -0.406347, 1
48.000000, 64.000000, 0.357806, -0.175874, 1
64.000000, 64.000000, 0.417865, 0.127232, 1
80.000000, 64.000000, 0.443664, 0.424766, 1
96.000000, 64.000000, 0.316776, 0.640383, 1
112.000000, 64.000000, 0.370178, 0.977602, 1
16.000000, 48.000000, 0.689415, -0.710014, 1
32.000000, 48.000000, 0.679637, -0.484498, 1
48.000000, 48.000000, 0.690547, -0.162003, 1
64.000000, 48.000000, 0.652830, 0.086346, 1
80.000000, 48.000000, 0.604724, 0.366265, 1
96.000000, 48.000000, 0.638212, 0.677957, 1
112.000000, 48.000000, 0.607792, 0.963164, 1
16.000000, 32.000000, 0.928225, -0.619016, 1
32.000000, 32.000000, 0.964546, -0.463650, 1
48.000000, 32.000000, 0.974076, -0.181383, 1
64.000000, 32.000000, 0.902069, 0.103916, 1
80.000000, 32.000000, 0.913123, 0.425505, 1
96.000000, 32.000000, 0.944750, 0.695440, 1
112.000000, 32.000000, 0.943867, 0.963157, 1
16.000000, 16.000000, 1.242310, -0.756268, 1
32.000000, 16.000000, 1.170329, -0.446783, 1
48.000000, 16.000000, 1.189673, -0.229518, 1
64.000000, 16.000000, 1.210342, 0.093494, 1
80.000000, 16.000000, 1.265605, 0.400713, 1
96.000000, 16.000000, 1.271305, 0.682317, 1
112.000000, 16.000000, 1.252162, 0.976547, 1
Velocity *.VEC File

A velocity *.VEC vector file is created by Insight software when an image is
processed in Velocity measurement mode. In this mode, the image field of
view is calibrated to convert from pixels to mm and m/s.

Velocity *.VEC Header Line
The following lists the information in the vector header file.
TITLE="\D:\Experiments\NoFlame\Vector\NoFlame00000.v Filename
ec"
VARIABLES= "X mm", "Y mm", "U m/s", "V m/s", "CHC" Data Columns Titles. Note that the vector positions are in mm
and the vectors are m/s.
DATASETAUXDATA Application="PIV" File calibration data. The application, X and Y pixel size in the
DATASETAUXDATA SourceImageWidth="512" fluid m/pixel, image origin, dT time between laser pulses in
DATASETAUXDATA SourceImageHeight="512" ms, and the image size in pixels.
DATASETAUXDATA
MicrometersPerPixelX="42.860001"
DATASETAUXDATA
MicrometersPerPixelY="42.860001"
DATASETAUXDATA LengthUnit="mm"
DATASETAUXDATA OriginInImageX="0.000000"
DATASETAUXDATA OriginInImageY="0.000000"
DATASETAUXDATA
MicrosecondsPerDeltaT="10.000000"
DATASETAUXDATA TimeUnit="ms" DATASETAUXDATA
SecondaryPeakNumber="0" DATASETAUXDATA
DewarpedImageSource="0"
Zone
I=5, J=5, Number of vector columns and rows.
F=POINT Data Record Format.
Velocity *.VEC File Listing

The following lists the information in the vector file.
TITLE="C:\Experiments7\Experiment PIV
Demo\Sim\Analysis\sim000001.T000.D000.P013.H001.L.vec" VARIABLES="X mm", "Y mm", "U m/s",
"V m/s", "CHC", DATASETAUXDATA Application="PIV" DATASETAUXDATA SourceImageWidth="512"
DATASETAUXDATA SourceImageHeight="512" DATASETAUXDATA MicrometersPerPixelX="42.860001"
DATASETAUXDATA MicrometersPerPixelY="42.860001" DATASETAUXDATA LengthUnit="mm"
DATASETAUXDATA OriginInImageX="0.000000" DATASETAUXDATA OriginInImageY="0.000000"
DATASETAUXDATA MicrosecondsPerDeltaT="10.000000" DATASETAUXDATA TimeUnit="ms"
DATASETAUXDATA SecondaryPeakNumber="0" DATASETAUXDATA DewarpedImageSource="0" ZONE I=8,
J=6, F=POINT
0.685760, -0.642900, -10.748755, -15.537544, 1
1.371520, -0.642900, -10.367813, -15.116728, 1
2.057280, -0.642900, -9.883249, -15.988159, 1
2.743040, -0.642900, 24.345598, 49.439709, 1
3.428800, -0.642900, -4.079507, 3.462374, 1
4.114560, -0.642900, -10.931281, -8.633723, 1
4.800320, -0.642900, -10.680726, -8.430496, 1
5.486080, -0.642900, -11.227771, -7.058886, 1
0.685760, -1.328660, -10.104179, -14.704104, 1
1.371520, -1.328660, -10.116139, -14.203558, 1
2.057280, -1.328660, -9.289243, -13.969359, 1
2.743040, -1.328660, -9.614551, -13.373343, 1
3.428800, -1.328660, -22.890966, -0.089707, 1

4.114560, -1.328660, -10.246766, -9.213731, 1
4.800320, -1.328660, -10.429434, -8.679404, 1
5.486080, -1.328660, -9.487607, -7.143462, 1
0.685760, -2.014420, -9.198798, -15.036974, 1
1.371520, -2.014420, -9.767220, -14.733873, 1
2.057280, -2.014420, -8.951193, -13.621287, 1
2.743040, -2.014420, -9.492549, -13.314558, 1
3.428800, -2.014420, -8.112553, -10.448337, 1
4.114560, -2.014420, -7.724287, -10.486112, 1
4.800320, -2.014420, -6.913253, -7.265481, 1
5.486080, -2.014420, -8.549618, -7.581446, 1
0.685760, -2.700180, -8.328514, -14.751987, 1
1.371520, -2.700180, -8.623814, -14.527966, 1
2.057280, -2.700180, -7.512438, -12.787611, 1
2.743040, -2.700180, -7.475470, -11.911947, 1
3.428800, -2.700180, -7.586742, -10.949389, 1
4.114560, -2.700180, -7.563378, -9.903362, 1
4.800320, -2.700180, -7.658662, -9.237724, 1
5.486080, -2.700180, -8.272442, -8.873545, 1
0.685760, -3.385940, -7.999977, -14.286297, 1
1.371520, -3.385940, -8.499722, -13.539602, 1
2.057280, -3.385940, -7.212042, -12.376672, 1
2.743040, -3.385940, -6.962935, -12.011209, 1
3.428800, -3.385940, -6.357596, -11.912609, 1
4.114560, -3.385940, -6.014137, -11.135893, 1
4.800320, -3.385940, -7.145750, -9.546227, 1
5.486080, -3.385940, -6.569643, -7.040615, 1
0.685760, -4.071700, -4.996195, -14.992633, 1
1.371520, -4.071700, -6.887574, -13.239851, 1
2.057280, -4.071700, -6.715889, -11.706197, 1
2.743040, -4.071700, -5.791245, -11.913688, 1
3.428800, -4.071700, -5.763193, -12.647861, 1
4.114560, -4.071700, -5.073339, -12.279845, 1
4.800320, -4.071700, -5.213373, -9.701417, 1
5.486080, -4.071700, -5.435574, -7.371693, 1
Average Vector (*AVG.VEC) File

The Average vector field file is created with Tecplot® software by averaging
a sequence of vector fields at each measurement location. This file format
is similar to the standard VEC file format. The changes are the velocity is
Mean Velocity and the fifth column is Count instead of Choice. The
Average Vector file can be included as part of an animation sequence in
Tecplot® software so that the animation can show a sequence of
instantaneous vector fields followed by the average vector field.
The count is the number of valid vectors (vectors with a positive choice
code) at each location. The maximum count is the number of vector fields
averaged together. If some vectors at a location were not valid, the count
at that point is reduced. Only valid points are included in the average.
The vector grid for all of the vector fields in the average must be the same.
Having a different number of vector columns and rows, or different vector

locations in some of the VEC files used in the average will give incorrect
results.
The units of the *AVG.VEC file may be in either pixels or mm and m/s
depending on the units of the VEC files used in computing the average.
Average Vector File Listing

The following shows an Average Vector file listing:
TITLE="\D:\Experiments\Deg\Vector\DegL” VARIABLES= "X pixel", "Y pixel", "U Mean pixel", "V
Mean pixel", "Count” ZONE T="Pixel, Height=1016, Width=1000 “ I=5, J=5, F=POINT
138.000000, 908.000000, 0.194089, 0.241566, 8
261.000000, 908.000000, 0.233431, 0.205879, 8
384.000000, 908.000000, 0.267045, 0.153003, 8
507.000000, 908.000000, 0.358548, 0.153543, 8
630.000000, 908.000000, 0.431884, 0.111491, 8
138.000000, 827.000000, 0.159519, 0.154415, 8
261.000000, 827.000000, 0.226142, 0.126557, 8
384.000000, 827.000000, 0.266780, 0.096698, 8
507.000000, 827.000000, 0.359502, 0.111918, 8
630.000000, 827.000000, 0.419891, 0.055146, 8
138.000000, 746.000000, 0.162204, 0.104445, 8
261.000000, 746.000000, 0.228636, 0.103401, 8
384.000000, 746.000000, 0.276518, 0.079071, 8
507.000000, 746.000000, 0.360926, 0.077221, 8
630.000000, 746.000000, 0.452794, 0.069008, 8
138.000000, 665.000000, 0.155498, 0.070431, 8
261.000000, 665.000000, 0.217330, 0.035222, 8
384.000000, 665.000000, 0.275173, 0.032714, 8
507.000000, 665.000000, 0.353082, -0.035741, 8
630.000000, 665.000000, 0.436182, 0.020788, 8
138.000000, 584.000000, 0.166035, 0.034134, 8
261.000000, 584.000000, 0.224729, 0.041150, 8
384.000000, 584.000000, 0.282870, 0.002764, 8
507.000000, 584.000000, 0.359145, 0.008448, 8
630.000000, 584.000000, 0.415600, -0.017352, 8
Mean and Standard Deviation (*.STD) Vector File

The mean and standard deviation file is computed by Tecplot® software
from a sequence of vector fields. The *.STD contains all of the data in the
*AVG.VEC file plus the standard deviation of the valid velocity
measurements at each location.
The *.STD file has two more data columns than the *AVG.VEC file and so
cannot be opened with a sequence of VEC files in Tecplot® software to
create an animation.
The units of the *.STD file may be in either pixels or mm and m/s
depending on the units of the VEC files used in computing the average.

Mean and Standard Deviation File Header
The following lists the information in the Mean and Standard Deviation File
header:
TITLE="D:\Experiments\Deg\Vector\DegLAvg.std" Filename
VARIABLES= "X pixel", "Y pixel", "U Mean pixel", "V Mean Data Column Titles
pixel", "Count", "U StdDev", "V StdDev"
ZONE T="Pixel, Height=1016, Width=1000 " Calibration data
I=5, J=5 Number of vector Columns and Rows
F=POINT Data Record Format
Mean and Standard Deviation File Listing

The following is a listing of the Mean and Standard Deviation file:
TITLE="D:\Experiments\Deg\Vector\DegLAvg.std” VARIABLES= "X pixel", "Y pixel", "U Mean

pixel", "V Mean pixel", "Count", "U StdDev", "V StdDev” ZONE T="Pixel, Height=1016,
Width=1000 “ I=5, J=5, F=POINT
138.000000, 908.000000, 0.194089, 0.241566, 8, 0.009636, 0.006454
261.000000, 908.000000, 0.233431, 0.205879, 8, 0.010775, 0.006944
384.000000, 908.000000, 0.267045, 0.153003, 8, 0.009808, 0.005208
507.000000, 908.000000, 0.358548, 0.153543, 8, 0.017558, 0.009246
630.000000, 908.000000, 0.431884, 0.111491, 8, 0.020065, 0.018837
138.000000, 827.000000, 0.159519, 0.154415, 8, 0.008751, 0.005492
261.000000, 827.000000, 0.226142, 0.126557, 8, 0.024162, 0.008655
384.000000, 827.000000, 0.266780, 0.096698, 8, 0.019269, 0.009730
507.000000, 827.000000, 0.359502, 0.111918, 8, 0.028638, 0.010139
630.000000, 827.000000, 0.419891, 0.055146, 8, 0.013219, 0.011245
138.000000, 746.000000, 0.162204, 0.104445, 8, 0.009589, 0.007419
261.000000, 746.000000, 0.228636, 0.103401, 8, 0.006463, 0.007079
384.000000, 746.000000, 0.276518, 0.079071, 8, 0.016142, 0.007830
507.000000, 746.000000, 0.360926, 0.077221, 8, 0.016967, 0.006335
630.000000, 746.000000, 0.452794, 0.069008, 8, 0.017017, 0.003942
138.000000, 665.000000, 0.155498, 0.070431, 8, 0.008919, 0.002538
261.000000, 665.000000, 0.217330, 0.035222, 8, 0.017503, 0.004883
384.000000, 665.000000, 0.275173, 0.032714, 8, 0.032399, 0.007115
507.000000, 665.000000, 0.353082, -0.035741, 8, 0.010044, 0.008536
630.000000, 665.000000, 0.436182, 0.020788, 8, 0.012876, 0.006777
138.000000, 584.000000, 0.166035, 0.034134, 8, 0.010848, 0.004685
261.000000, 584.000000, 0.224729, 0.041150, 8, 0.013074, 0.002998
384.000000, 584.000000, 0.282870, 0.002764, 8, 0.015167, 0.004134
507.000000, 584.000000, 0.359145, 0.008448, 8, 0.021311, 0.002282
630.000000, 584.000000, 0.415600, -0.017352, 8, 0.030514, 0.005010

Particle (*.PAR) Files
The Particle file is created by Insight software when an image is processed
with PTV processor (.PAR) or GSV sizing processor (.GSVX.PAR). Particle
files may be in units of mm and m/s, or in pixels depending on the
calibration mode selected. The file header line labels the data columns and
shows which measurement units were used. In addition to the particle
position, velocity and size information, a particle choice code value is
recorded in the particle record to show the status of the particle.
Choice (CHC) Codes

The choice code in the particle record identifies some information about the
particle. Only particles with non-negative CHC codes should be included in
the final particle field, and only particles with positive CHC codes should be
accepted in velocity field computations. The following lists the choice
codes.
Table A-4
Choice Codes
CHC Description
1 Particles with a valid velocity vector
0 Particles obtained without velocity processing
-3 Particles with an invalid velocity vector
-4 Particles without velocity vector
Velocity Not Set

Velocity values of 9.9e9 indicate the particle velocity has not been set. It
can happen when no velocity processing has been done for the particles.
For example, only particle identification is done in PTV processor, or only
GSV sizing is done in GSV. The choice code 0 is used in these cases.
Velocity Not Set can also happen when no velocity is found for some
particles in the velocity processing. The choice code -4 is used for those
particles.
Scalar Data (*.SCA) Files

The Scalar Data file is created by Insight scalar data export processor,
which is available in the post-processing of PLIF application. Scalar data
file may contain single-camera PLIF result, simultaneous PLIF-PIV result,
or multi-camera PLIF result.

Single-Camera PLIF
The space unit may be in units of pixels or mm for single-camera PLIF,
depending on whether the spatial calibration is provided during scalar data
export.
Table A-5
Single-Camera PLIF
Columns Description
1 Grid point X
2 Grid point Y
3 Choice code
4 Scalar data
Table A-6
Choice Codes
CHC Description
1 Grid points with a valid scalar
-3 Grid points excluded from PLIF processing
Simultaneous PLIF-PIV
The Scalar Data file for simultaneous PLIF-PIV contains the scalar data
from the PLIF camera and the vector data from the PIV camera. The space
unit is always in mm because a perspective calibration is required when
combining the PIV and PLIF results during Scalar Data Export.
Table A-7
Simultaneous PLIF-PIV
Columns Description
1 Grid point X
2 Grid point Y
3 Choice code
4 Scalar data
5 Grid point X in pixels
6 Grid point Y in pixels
7 Vector X
8 Vector Y

Table A-8
Choice Codes
CHC Description
1 Grid points with a valid scalar.
-3 Grid points excluded from PLIF processing or
without valid PIV vector.
Multi-Camera PLIF
The Scalar Data file for multi-camera PLIF contains the scalar data from
multiple PLIF cameras. The space unit is always in mm because a
perspective calibration is required when combining the PLIF results from
multiple cameras during Scalar Data Export.
Table A-9
Multi-Camera PLIF
Columns Description
1 Grid point X
2 Grid point Y
3 Choice code
4 Scalar data, camera 1
5 Scalar data, camera 2
… …
Table A-10
Choice Codes
CHC Description
1 Grid points with a valid scalar.
-3 Grid points excluded from PLIF processing or
without valid PIV vector.
Statistics List (*STL) File

The Statistics List File gives a summary of the vector file data for a
sequence of vector fields. The first line is the header that labels the data
columns. The Statistics List File can be read into a spreadsheet program.
Viewing this file can show how a sequence of vector fields are changing.
For example, did some vector fields require more editing than the others,
how are the mean velocity values changing over time?
To Read the Statistics File into Microsoft® Excel® Spreadsheet

Program
1. Click File | Open to get the file selection dialog box.
2. Change the Files of Type to All Files (*.*).
3. Select the Statistics (*STL) file.

4. Using the Text Import Wizard dialog, select Delimited.
5. Click Next.
6. On the next page select Comma.
7. Click Finished. The data is read into an Excel® spreadsheet with the
column titles in Row A, and one line for each file in the list.
Statistics File Header

The following table shows the header of the Statistics file.
File Filename
Velocity Statistics
U Mean Average U velocity values in vector field
V Mean Average V velocity values in vector field
U StdDev Standard Deviation of U velocity
V StdDev Standard Deviation of V velocity values in vector field
Velocity Range
U Max Largest U velocity value
U Min Minimum U velocity value
V Max Largest V velocity value
V Min Minimum V Velocity value
Choice Code Counts
1st Choice Number of first peak vectors, choice code 1
2nd Choice Number of second peak vectors, choice code 2
3rd Choice Number of third peak vectors, choice code 3
Smoothed Number of measured smoothed vectors, choice code 5
Interpolated Number of interpolated vectors, choice code 4
Temp Blank Number of temporally blank vectors, choice code 0
SNR Fail Number of vectors where correlation failed SNR criteria ,
choice code -1
Removed Number of vectors removed by polygon edit, choice code -2
Vector Units
Velocity Unit Vector field units either Pixels or m/s
Statistics File Listing

The following is an example listing of a statistics file:
File, U Mean, V Mean, U StdDev, V StdDev, U Max, U Min, V Max, V Min, 1st Choice, 2nd
Choice, 3rd Choice, Smoothed, Interpolated, Temp Blank, SNR Fail, Removed, Velocity Unit
d:\experiments\deg\vector\deg00000l.vec, 0.284133, 0.081017, 0.098876, 0.067917, 0.460182,
0.150314, 0.236902, -0.025677, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.156136, 0.231293, -0.030029, 25, 0, 0, 0, 0, 0, 0, 0, pixel

0.142189, 0.240892, -0.042515, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.161285, 0.243132, -0.049000, 25, 0, 0, 0, 0, 0, 0, 0, pixel
d:\experiments\deg\vector\deg00006r.vec, -0.292504, 0.135428, 0.120730, 0.070908, -
0.123068, -0.507261, 0.273191, 0.013832, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.133028, -0.478882, 0.272966, 0.011860, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.129911, -0.468920, 0.264456, 0.009594, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.133997, -0.477514, 0.267853, 0.017067, 25, 0, 0, 0, 0, 0, 0, 0, pixel
0.149426, 0.241810, -0.029633, 25, 0, 0, 0, 0, 0, 0, 0, pixel
d:\experiments\deg\vector\deglavg.vec, 0.056147, 0.101641, 0.101377, 0.065080, 0.207377, -
0.099389, 0.232638, 0.008464, 25, 0, 0, 0, 0, 0, 0, 0, pixel
d:\experiments\deg\vector\deglAVG.vec, 0.032628, 0.103661, 0.101940, 0.065097, 0.183320, -
0.125280, 0.231745, 0.011045, 25, 0, 0, 0, 0, 0, 0, 0, pixel
Image Capture Timing (*.tstmp) File

When a sequence of images is captured to RAM, a timing file is stored in
the experiment folder showing the times when the images were captured.
All times recorded in this file are in milliseconds.
The first line is the time and date of the opening of the log file.
The image filename is shown in the first column.
The second column is the elapsed time since the beginning of the capture.
Note that the elapsed time between the beginning of the capture and the
first actual capture will always be larger than the time between subsequent
captures. This is due to various startup activities (memory allocation,
synchronizer communication, etc.) that take place during this time.
The third column shows the time since the last frame captured and can be
analyzed to look for missed triggers. To get the frame rate in frames per
second, use 1/time/1000.
The fourth column shows the time spent solely in the frame grabber. In
certain cases this can be used to determine if other processes are affecting
the capture rate.
Timing File Listing

The following is an example of a timing (*.tstmp) file listing the camera
timing information.

Figure A-1
Timing File Listing
3-D Vector Files

The 3-D vector file is created by combining two, 2-D vector files into one
3-D vector file. The file contains a header line and one record for each
vector.
The U, V, and W velocity components are computed using at least-squares

fit of the left vector and right vector. The residual error found by plugging
the least-squares solution back into the equations. The total residual error
is the square root of the component residual errors squared. The units of
the residual error are in pixels. For example if the residual error is 1.0 pixel,
the left and right vectors did not match by one pixel. In typical PIV
configurations this indicates that the two V velocity measurements differed
by 1 pixel. The residual can be used as a check to make sure that the two
2-D vector can be from the same 3-D vector.
The choice codes are listed in the VEC file format definition. Non-positive
choice codes indicate errors and must not be used in velocity field
statistics.
3-D Vector Header Line

Following shows a 3-D vector file header:
TITLE="D:\Experiments\Deg\Vector\Deg00000.v3D" filename
VARIABLES="X mm", "Y mm", "Z mm", "U m/s", "V m/s", Data Column Titles
"W m/s", "CHC", "Residual pixels"
ZONE T="3D Velocity" Title
I=4, J=4, K=1 Columns, Rows, Planes, K is always 1 for stereo PIV
F=POINT Record Format

Note: The V3D file units are always in mm and m/s, not pixels.
3-D Vector (*.V3D) File Listing

The following is a 3-D Vector file listing:
TITLE="D:\Experiments\Deg\Vector\Deg00000.v3D” VARIABLES="X mm", "Y mm", "Z mm", "U m/s",

"V m/s", "W m/s", "CHC", "Residual pixels", ZONE T="3D Velocity” I=4, J=4, K=1, F=POINT
-9.60459, -15.9296, 0, -0.000524411, 0.00414582, -0.461622, 1, 0.0314854
-6.20357, -15.9296, 0, 0.000194652, 0.00430889, -0.442992, 1, 0.0159115
-2.80255, -15.9296, 0, -0.000152724, 0.00397104, -0.420229, 1, 0.0291489
0.598478, -15.9296, 0, 4.18488e-005, 0.00395065, -0.440376, 1, 0.0357319
-9.60459, -12.5285, 0, 9.20153e-005, 0.00487148, -0.453866, 1, 0.00915304
-6.20357, -12.5285, 0, -3.58881e-005, 0.0044658, -0.45274, 1, 0.0157696
-2.80255, -12.5285, 0, 0.000419979, 0.0045025, -0.419085, 1, 0.0297433
0.598478, -12.5285, 0, 0.000382887, 0.00470845, -0.427999, 1, 0.00732069
-9.60459, -9.12751, 0, 0.000157937, 0.00271535, -0.46542, 1, 0.0158749
-6.20357, -9.12751, 0, -0.000528862, 0.00380699, -0.42884, 1, 0.034495
-2.80255, -9.12751, 0, -0.0016384, 0.00325252, -0.385416, 1, 0.0543617
0.598478, -9.12751, 0, -0.000608029, 0.00317918, -0.405632, 1, 0.0737102
-9.60459, -5.72648, 0, -0.000905317, 0.00278981, -0.476492, 1, 0.0034331
-6.20357, -5.72648, 0, -0.0002381, 0.00213592, -0.430195, 1, 0.0281619
-2.80255, -5.72648, 0, 0.000811252, 0.00254835, -0.450617, 1, 0.00909315
0.598478, -5.72648, 0, 0.00182587, 0.00309806, -0.413516, 1, 0.0262402


APPENDIX B
Using MATLAB®
Program
MATLAB®, a technical computing program combining computation,

programming, and visualization, is integrated into Insight™ software.
Specifically, the Spatial Tool, Time Series Analysis Toolbox, and Proper
Orthogonal Decomposition (POD) features can be accessed and used from
within Insight software to perform PIV data analysis and to display PIV
results such as advanced statistical properties in time and space domains.
Getting Started
To access these features, select Tools | MATLAB | Spatial and Time
Series Analysis… or Proper Orthogonal Decomposition from the
Insight main menu.
Figure B-1
Spatial Toolbox Window
MATLAB® is a registered trademark of The MathWorks, Inc.
B-1
Loading Insight Software Files
Insight™ data, vector files with *.vec or *.v3d extensions can be loaded into
this program.
To load Insight data files:

1. From the main menu bar of the Spatial Toolbox screen, click File |
Load Vector Files. A dialog box to select vector files opens.
Figure B-2
Select Vector Files
2. Use of the following methods to select and load the files:
 Click to browse to the appropriate folder and select the folder

that includes the data files.
 If you know directly type in the name of the folder.
 Double-click the .. in the list of files box, or double-click the name
of the directory within the list to get the files in the subfolders. *.vec
files are listed as a default. Check Switch to .v3d to see the list of
*.v3d files.
Note: The order in which files are selected is not important. You can
pick the last file first, and the first file last. During the loading
procedure, the files are automatically sorted in the ascending
order of their filenames.
 Set dT for the time interval between two vector files.
B-2 Insight 4G Global Image Acquisition, Analysis and Display Software

3. Click Load. The Spatial Toolbox reads the vector files, initializes the
toolbox default variables and calculates several flow quantities. This
part may take from several seconds to a few minutes, depending on
the length of the selected series and the size of each vector map.
Using the Spatial Toolbox

When you select at least one vector file in the Spatial Toolbox, the default
window looks like the one shown in Figure B-3:
Figure B-3
Vector File Selected
Using MATLAB® Program B-3

Navigating the Spatial Toolbox
 Use Next or Previous to display
flow fields maps. The appropriate
map number is also displayed.
 To jump to specific map, enter the
number of the map in the edit box
and click Enter.
Spatial Toolbox Options

Choose to display data using one large set of options with various
combinations and choose arrows either in black or in color. The color of
arrows could represent any one of the available quantities for that
particular mode. The arrows could be on top of the contour map of the
selected quantity, and contour could be displayed as color patterns,
bounded, or unbounded (smooth view), color or black contour lines. The
following lists the control options available in the Spatial Toolbox.
Arrow Options
The following arrow choices are available:
Option Description
On/Off Select On to display and Off to hide the arrows.
Color/Black Available only if arrows are turned on.
Color: Displays arrows according to the current
quantity.
Arrow Size Specify arrow size.
Contour Quantity Modes

Use this option to choose the quantity that will be displayed. There are four
main sets of quantities or modes.
To Choose Contour Quantity Mode
Use the two checkboxes to switch between the

following modes:
Instantaneous Mode
Leave both checkboxes unselected to choose Instantaneous mode The list
of quantities is:

Ensemble Averaged Mode
Select the Ensemble checkbox to choose Ensemble Average mode. This
gives the ensemble averages of the instantaneous quantities. The list is:
Fluctuations Mode
Select the Fluctuations checkbox for this mode. The list is:
Turbulent Mode
Select both checkboxes for Turbulent quantities mode. The list is:
Contour Types
To specify a contour type, choose one of the following options in the
Contour Type box:
Option Description
Flood Displays smooth color patches with no sharp
boundaries.
Color Line Displays only color boundaries, according to the color
map of the selected quantity.
Flood + Line Displays color patches, with sharp boundaries.
Black Line Displays only boundaries of the contours.

Defining Contours
Setting the number of colors, using either default settings or by specifying
them manually, defines the number of contour levels of a selected quantity,
in one of the following four contour modes:
Mode Description
Each Field Select to automatically update the contour level to each
presented map limits (min and max) and distributed evenly
into number of contour levels, that is number of colors.
All to Display Select to have the currently presented map define the
contour levels for all following maps, unless another option is
selected. If you change the quantity, the limits are updated,
but remain constant for all forthcoming views.
All Fields Select to define the contour levels by the maximum and
minimum values of the selected quantity from the whole
selected dataset. This is a powerful option allowing you to
compare the values visually, with not a single value out of the
defined color set (which can happen in the Manual or All to
Display modes). However, to determine the absolute
maximum and minimum values, the entire dataset has to be
calculated at least once. For example, All Fields of the
vorticity demands the calculating gradients in both directions
of the selected dataset. This could be time-intensive for the
time-resolved PIV images).
Manual Specify the maximum and minimum level of the color to be
shown, the selection could be symmetric by entering the
same positive and negative values or anti-symmetric, if
different values are used.
To Specify Number of Colors (Contour Levels):

1: Enter the desired number in Number of Colors edit box:
2: Click Enter.

Figure B-4 shows the instantaneous velocity field (arrows) and the
instantaneous vorticity field in color-line contour mode. Color bar is added
and the number of contour levels is set to 25.
Figure B-4
Instantaneous Velocity Field (Arrows) and the Instantaneous Vorticity Field in
Color-Line Contour Mode

Figure B-5 shows an instantaneous velocity field with longer arrows, set to
3. Flood contour type is of the rate-of-strain component, the number of
contour levels are set to 3 to emphasize strong positive and negative
regions.
Figure B-5
Instantaneous Velocity Field with Longer Arrows

Figure B-6 shows an ensemble averaged velocity field displayed with
arrows. The color of the arrows and the sharp boundaries are of the
average vorticity field, the number of contour levels is set to 5, and the
color is distributed symmetrically by manual set of contour levels between
–0.04 and 0.04 [1/sec].
Figure B-6
Ensemble Averaged Velocity Field Displayed with Arrows

Animating Displays
Run the animation of the successive maps, with all the visualization
properties that were selected before, using the animation feature of the
Spatial Toolbox.
Note: During animation, the movie is not stored in the memory, you need
to explicitly save it using the Movie option. See “Creating Movies.”
To Start Animation
Click Animate. The first map is the current map (that could be manually
entered in the edit box).
To Stop Animation
Click Animate again. The animation stops at any map when the action was
performed. If you do not stop, the animation continues until the last map is
displayed.
Creating Movies
The movie option allows you to save your animated maps in the form of an
.avi file.
To Create a Movie
1. Click Movie.
2. Enter a name for the .AVI file. The Windows AVI video file
(uncompressed, default frame-per-second rate of 15 fps) is saved in
the Insight directory, when the last map is reached, or when the Movie
button is clicked again.

Using the Time Series Analysis Toolbox
Use the Time Series Analysis Toolbox feature of MATLAB® program to
make selections such as points or regions in the images and then use
various options to get a profile or time distribution analysis.
To Access the Time Toolbox

Click >Select<.
A collection of selection tools become available.
To Make a Selection
Click the appropriate selection. These are described later in this section.
Note: You can select only one type of selection. For example Add Points
cannot be selected with Add Region.
To Quit After Making a Selection

After region/points are selected, click the middle mouse button.
To Reset a Selection
Click Reset. All selections are cleared. Also when you click Spatial, all
selections are cleared.

Selecting Options for Analysis
The following describe each of the selection options that are available.
Selecting Points
This selection allows you to select single/multiple points from the map as
shown in Figure B-7.
To Select Points
Drag and select a region and then click Select/Add Points.
Figure B-7
Select Single/Multiple Points from the Map

Selecting Regions
This selection allows you to select particular regions in your map for
analysis.
Note: When adding two or more regions, you can select only regions that
have at least one side equal, and ends of the side that lie on the
same line.
To Select Regions
Drag the mouse and select a region and then click Select/Add Regions.
Figure B-8 shows two regions selected.
Figure B-8
Two Regions Selected

Selecting Rows or Columns
These options allow you to select rows and columns on the map
(Figure B-9).
To Select Rows or Columns

 Click Select/Add Row to select the entire row.
 Click Select/Add Column to select the entire column.
Figure B-9
Selecting Rows (left) or Columns (right)

Selecting the Entire Map
This selection allows you to select the entire map.
To Select the Entire Map

Click Select All to select the entire row, as shown in Figure B-10.
Figure B-10
Select All to Select Entire Row

Performing Profile Analysis
After selecting areas of interest in the display, you can perform Profile
analysis.
To Perform a Profile Analysis

1. Select an area or point.
2. Click Profile. The distribution window opens.
3. Use the following options to view the display or export the information.
Option Description
Direction Changes one of the axis to x or y, respectively.
(Rows/Columns
Swap X-Y Changes between axes.
Single Displays all the data in the selected region of interest
(rows or columns or regions or all) in spatial
distribution along rows or along columns (X or Y
profile per row or column).
Average Displays red line presenting average of quantities.
Export Figure: Allows you to use the Plot Edit options of
®
MATLAB program.
CSV: Allows you to export the selected filename,
for saving data in CSV format.

Examples
This section provides screen shots illustrating the outputs of the various
options you may perform.
Figure B-11 shows a distribution window, with the Average distribution
added to the plot (thick red line) and displaying two disconnected regions.
Figure B-11
Distribution Window with Average Distribution Added to Plot and Displaying Two
Disconnect Regions

Figure B-12 shows distribution with axes swapped (rotated 90 degrees) by
checking the Swap X-Y option.
Figure B-12
Distribution with Axes Swapped

Figure B-13 shows the output of using the Export to Figure option.
Figure B-13
Output Using the Export to Figure Option

The Export to Figure allows you to copy one curve from the figure into
another, add text, arrows, etc., and change the visual style of the figure, as
shown in Figure B-14.
Figure B-14
Change the Visual Style of the Figure

Performing the Time Series Analysis
Use the Time Series tools to analyze high frame rate PIV images in the
same way as the other images.
To Perform Time Series Analysis

1. Select a region as described in “Selecting Options for Analysis” and
shown in Figure B-15. You could select a set of points, rows, columns,
regions or the entire map.
Figure B-15
Select a Region
2. Click Time Analysis.

Figure B-16 shows the first window of the Time Series Toolbox. The default
plot is spatial averaged time history of the u velocity (from left to right).
Figure B-16
First Window of the Time Series Toolbox
Other possible selections could be used for velocity components such as

instantaneous and fluctuating velocity components from the left to right (u),
from the top to bottom (v) and in the case of stereoscopic PIV data, the
out-of-plane component (w).

Figure B-17 shows the time history of a selected point, with the Time
Series option checked.
Note: The Average option is checked too, and it is emphasized by the thick
blue line on the display.
Figure B-17
Time History of a Selected Point
Depending on the two checkboxes, choose to see only pointwise time

history and/or its spatial average. In addition, you can plot the time history
of the fluctuation velocities, same as the instantaneous for the three
components.

Figure B-18 shows an example of the lateral correlation. Rii relates to the
selected velocity component (that is if u is selected it is Ruu, from the left to
the right) and k1 refers to the rows direction (from the top to the bottom).
Figure B-18
Example of the Lateral Correlation

Figure B-19 shows an example of the longitudinal correlation function, in
this case it is Rvv in k2 direction.
Figure B-19
Example of the Longitudinal Correlation Function
Figure B-20 shows an example of the wave-number spectra.
Figure B-20
Example of the Wave Number Spectra

Figure B-21 shows how each one of the axes could be independently
exported or saved into MATLAB® figure or comma separated file.
Figure B-21
Export to MATLAB® Program

Figure B-22 shows the Export to Figure option used. This option allows you
to use Plot Edit options of MATLAB® program.
Figure B-22
Export to Figure Option

For each comma separated file, a unique filename is used.
Figure B-23
Input Name for CSV

Using the Proper Orthogonal
Decomposition (POD) Toolbox
When at least one vector file is selected, the select window appears
allowing you to select region of interest.
Figure B-24
Default Selection Window

Select Options
In selection tab, the following selection tools are present (see Figure B-25).
Figure B-25
Select Options
Select/Add Region
Select a single region from the map (see Figure B-26).
Figure B-26
Selecting Regions

Select All
Pressing Select All will select the entire map (see Figure B-27).
Figure B-27
Selecting the Entire Map
When selection of region or selection of entire map is done, the only choice
is to reset current selection and to do another selection, or to continue with
the current.
Resetting of the current selection can be done

by pressing the Reset button.
Direct/Snapshot
This checkbox is checked and unchecked
automatically by the software. There is still an
option for the user to change the selection; however, the performance of
the toolbox will be strongly altered. There are two methods to calculate the
POD modes, as it is explained in details in theoretical and computational
background: the direct method and the method of snapshots. The selection
is done according to the minimal size of the input matrix, i.e., if the number
of vectors is less than the number of the velocity vector fields, then the

computationally efficient method is the direct method. Otherwise, the most
efficient method is the method of snapshots.The recommendation is not to
change the default selection of the software. Both methods lead, obviously,
to the identical result.
After selection region of interest, you may

proceed to choose the energy content by
pressing the Start button.
Or to adjust the selection, press Reset button.
Choosing Energy Content

After pressing the start button, plot of relative energy versus number of
modes is shown in Figure B-28.
Figure B-28
Plot of Relative Energy Versus Number of Modes

The following controls are enabled (Figure B-29) to let you choose number
of modes or relative energy percent. There are two choices:
 Enter number of modes
(minimum is 1, maximum is
number of maps loaded) and
press Continue button. Related
energy percent will be
automatically calculated.
 Enter energy percent (1..100%)
and press the Continue button.
Related number of modes will
Figure B-29
be automatically calculated. Energy Content Controls
Energy Content Controls

Energy plot can be switched to log-log (Figure B-30) notation using the
checkbox in plot options field (Figure B-31).
Figure B-30
Energy Plot Using Log-Log Notation

Energy plot can be saved to CSV file using Save to CSV button
(Figure B-31). When pressed, you will be prompted for the filename
to save.
Figure B-31
Plot Options Field and File Name Dialog
After pressing the Continue button, default window looks like Figure B-32.
Figure B-32
Default Window, Arrows, GUI Controls, and Main Axes

Navigation
Navigation is done by using the
Next/Previous buttons, which
display next/previous maps
accordingly.
To jump to a specific map, enter

the number of the map in the edit
box followed by Enter.
Available Options
Choose between one of the large set of various combinations to present its
data: arrows in black or in color. The color of arrows could represent any
one of the available quantities for that particular mode. The arrows could
be on top of the contour map of the selected quantity, and contour could be
as color patterns, bounded, or unbounded (smooth view), color or black
contour lines. Below, in the working example, we show some of the
combinations.
Arrow Options
The following arrow choices are available:
Option Description
On/Off Select On to display and Off to hide the arrows.
Color/Black Available only if arrows are turned on.
Color: Displays arrows according to the current quantity.
Contour Quantity Modes

Use this option to choose the quantity that will be displayed (see
Figure B-33).
The list of quantities is:
v u u u v v u v u v
u, v, u 2  v 2 ,    , s xx  , , , s yy  ,  , s xy  
x y x y x y x y y x

And in the same order they appear in the toolbox:
Figure B-33
List of Available Quantities to Use for a
Contour- or Color-Coding of an Output
Output Options
The results can be presented in Single POD modes, Multi-mode, or
Reconstruction of fields (Figure B-34).
Figure B-34
List of Available Ways to Show the
Output
Single POD Modes

Every POD mode of velocity vector fields could be shown as a vector field,
by arrows, and with derived quantities as the background contours. The
derived quantities are calculated instantaneously from each POD mode, as
it has been a regular flow snapshot. Navigation and animation/movie
options are: Next/Previous (from one to the number of modes, selected
before) and Animation/Movie from the current mode towards the last.
Multi-Mode
The result is a linear combination of POD modes, defined by the user in the
vector of indices of modes. The combination could be of consequent or
randomly selected modes (i.e., from 1 to 5 or 1,2,5, for example). In
addition, the toolbox allows calculating the sum of the selected modes or
their linear combination by using their respective eigenvalues. The latter
option is called weighted version and realized by checking the radio-button
(the short version of Weighted/Non-Weighted). The default is a
regular sum (Non-weighted version) and checking the radio-button will

show the weighted version. When working in Multi-mode, the control
shown in Figure B-35 is available.
Figure B-35
Multi-Mode Controls
The edit-box allows you to choose modes that will be used to

calculate quantities in multi-mode.
The software allows you to use any index notation of a vector in MATLAB®
program and gives lot of flexibility, allowing to choose single modes, group
of modes, group of modes with given step, etc. For example:
 1:10—Choose all modes from 1 to 10.
 1:2:10—Choose all modes with step of 2, i.e., 1,3,5,7,9 will be chosen.
 1, 3, 5 or 1 3 5—modes 1, 3, and 5 will be chosen and any other
combination, like 1:2:5,10 will also work
Note: Minimum allowed value to enter is 1 and maximum allowed value is
maximum number of modes.
Reconstruction of Fields
Every one of the selected vector fields could be reconstructed by using all
the modes or some of the modes. The reconstruction with all the modes
gives the exact original vector field, while the reconstruction with lower
number of modes will provide you with the low-order reconstructed model
of the vector field. You are advised to understand more about the low-order
representation from additional documentation providing a theoretical and
computational background.
When working in Reconstruction mode, the option shown in

Figure B-36 becomes available.
Figure B-36
Reconstruction Mode
You can make custom selection of modes to use in reconstruction, similarly

to the way explained in the multi-mode section.
Note: A default selection for reconstruction and multi-mode is a vector of

[1:maximum number of modes].

Contour Types
Available contour types are:
Flood Smooth color patches, no sharp boundaries.

Color Line Only color boundaries, according to the color map of
the selected quantity.
Flood + Line Color patches, with sharp boundaries.
Black Line Only boundaries of the contours.
There are several examples given below to show some of the available
arrow and contour combinations.
Figure B-37
Single Modes Velocity Field (Arrows) and the Instantaneous Vorticity Field in Color-
Line Contour Mode. Note that a number of contour levels are set to 25
Control of Contour Levels and Colors

To manually enter the number of contour
levels (or number of colors), enter the
desired number in the edit box and
pressing Enter.

The number of colors (default or manually set) defines the number of
contour levels of the selected quantity, but in one of four modes (contour
modes):
Each Field The contour is automatically updated to each

presented map limits (min and max) and distributed
evenly into number of contour levels (i.e., number of
colors).
All to Display The currently presented map defines the contour levels
for all forthcoming maps, unless other option is
selected. If you change the quantity, the limits are
updated, but still will be constant for all forthcoming
views.
All Fields The contour levels are defined by the maximum and
minimum values of the selected quantity from the
whole selected dataset. This is a very powerful option
that allows you to compare the values visually, when
not even one value will be out of the defined color set
(which is possible to happen in manual or All to Display
modes). However, you should realize that in order to
determine the absolute maximum and minimum values,
the whole dataset has to be calculated at least once.
For example, All Fields of the vorticity demands the
calculation of gradients in both directions of the
selected dataset.
Manual Select the maximum and minimum level of the color to
be shown, the selection could be symmetric by entering
the same positive and negative values or anti-
symmetric, if different values are used.

Figure B-38
Multi-Mode Output, Non-Weighted Sum of the First 5 POD Modes. Contour is set to
flood+line type of the rate-of-strain component syy, number of contour levels set to 6
to emphasize strong positive and negative regions.
Animation and Movie Options

Run the animation of the successive maps, with all the visualization
properties that were selected before, by pressing Animation button. The
first map is the current map (that could be manually entered in the edit box)
and pressing again the Animation button stops the animation at any map.
Otherwise, the animation will continue up to the last map.
During the animation, the movie is not stored in the memory. You should
operate the Movie button in the same way as the Animation button. But in
addition, you will be prompted to type in the file name of the AVI file. The
Windows AVI video file (uncompressed, default rate is 15 frames-per-
second) is saved when the last map is reached, or the Movie button is
pressed again (i.e., released).
Note: The POD Toolbox window must be in the front while the movie is
recorded. The movie is recorded by using MATLAB® standard
function getframe() that captures the screen. Therefore, if
another window pops up in front of the POD Toolbox, it might be
captured and it may corrupt the movie.

Figure B-39
The Color of the Arrows and the Contour Lines Correspond to Vorticity, the Number
of Contour Levels is set to 6, and the Color is Distributed Symmetrically by Manual
Set of Contour Levels Between –0.001 and 0.001


APPENDIX C
Enabling and Setting
the Traverse
This appendix describes how to enable and setup a traverse, if you are
using one in your experiment.
The traverse manager is used to control the position of the laser/camera. It
has two modes:
 Manual Move Mode in which the motion of the laser or camera is
controlled manually and not by Insight™ software.
 Traverse Manager Move Mode in which motion is controlled by the
traverse manager).
Traverse control can be enabled by selecting Tools | Hardware Setup |

Traverse Installed. The position values can be displayed by clicking
Traverse Setup.
Figure C-1
Traverse Manager Dialog Bar
To access the Traverse Manager click the Setup button under Traverse
Setup.
C-1
Enabling and Setting up the Traverse
To Enable and Setup the Traverse
1. From the Capture tab, click Component Setup. Select the Traverse
Manager tab in the dialog box.
2. Click OK.
3. From the Capture tab, click Traverse Setup. The traverse setup box
appears.
Figure C-2
Traverse Manager Screen
C-2 Insight 4G Global Image Acquisition, Analysis and Display Software

To Change to Manual Move Mode
1. Click Setup in the Traverse Manager dialog box. The Traverse Setup
Dialog appears.
Figure C-3
Traverse Setup –
Communication
Setup Screen
2. Select the Manual Move box. When Manual Move is selected, all other
settings are unavailable.
To Change to Traverse Manager Control

1. Click Setup in the Traverse Manager dialog box. The Traverse Setup
Dialog appears.
2. Unselect the Manual Move box.
When Manual Move option is unselected, the driver, used to
communicate with a specific motion controller and the communications
port setting can be changed. The OK and Apply buttons apply the
changes made in the Communication Setup. The Cancel button
discards all changes made in that session.
Enabling and Setting the Traverse C-3

To Set Up and Configure Axis
1. From the Traverse Setup dialog box, click Axis Setup tab. The Axis
Setup dialog appears:
Figure C-4
Traverse Setup –
Axis Setup
Screen
2. Select an axis.
3. To enable the axis select the Axis Enable box. The following options
become available. Make your selections based on the following
descriptions.
Option Description
Title Title of the axis to be displayed.
Units Set the unit of traverse move.

Multiplier Set the multiplier of the steps per unit.
Digits Set the total number of floating point digits to be
displayed.
Time Out Set the timeout of each axis for traversing. The
software will stop the move and give a timeout.
Velocity (Steps) Set the velocity setting of the axis.
Acceleration (Steps) Set the acceleration setting of the axis.
4. Select Enabled under Relative Home to enter a number to denote a

home position relative to the hardware home position in the Position
box.
5. Click Apply and then OK.

To Use the Traverse Manager
Option Description
Move Sets the Move positions. The Move button moves all
enabled axis to the location specified in the edit box.
Jog The Jog positions of the X, Y, Z, U, V, and W axis. Moves
relative to the current position in the specified number of
steps.
+ - The + and - buttons within each axis group, jogs only that
axis.
Mode Switches communications mode with the motion
controller.
Local: Changes the positions outside of the application
usually through an external motion controller interface
(that is, joystick or front panel keypad).
Remote: Commands are sent through the serial port to
the motion controller.
An example of switching between communications mode may be that in

the current setup, the laser/camera position is off of the "home” position
needed:
 You select local mode and move the laser/camera using a joystick
through the motion controller.
 When you have the position needed, click the SetRel0 button to set the
relative home.
 Reselect remote mode.
The Rel Home button moves the laser/camera to the set relative home
position. The Abs Home button moves the laser/camera to the hardware
home position.
When an experiment is saved, the current location of the laser/camera is

saved. When an experiment is reloaded and run, the traverse will move to
the positions previously saved in the experiment file and the experiment
will continue.
If the experiment is run in Manual mode, a dialog box displays requesting

you to move to specified position.
Figure C-5
Manual Move Screen
Enabling and Setting the Traverse C-5


Index
autocorrelation at two points in calibration (continued)

1 turbulent flow, 10-8 Scheimpflüg angle, 9-8
12-bit images, 9-24
automatic range, 16-30 single-plane fiducial marker, 9-16
average field, 16-15 target, 9-5
2 average intensity, 8-18 target definition, 9-15
2-D calibration, 9-1 average intensity image, 15-21, 15-22 target location, 9-37
2-D particle image velocimetry, 8-1 average option, 16-15 two-plane fiducial marker, 9-18
2-D spatial calibration, 8-2 average spot intensity, 10-22 calibration analysis tools, 9-22
screen, 9-2 average vector file *AVG.VEC file calibration file
format, A-8 creating, 9-27, 9-30
3 average vector file *AVG.VEC file viewing, 9-30
3-D vector (*.V3D) file format, A-16 listing, A-9 calibration image analysis
3-D vector (*.V3D) file listing, A-17 axis enable, C-4 dialog, 9-12, 9-14
3-D vector fields, 8-1 axis setup, C-4 calibration images, 7-14, 9-10
3-D vector header line, A-16 calibration marks, 9-14
B calibration measurements
6 background image subtraction, 8-17 through a window, 9-47
630090 4MP-LS camera, 14-1 configuring preprocessor, 8-17 calibration points, 9-16
using two preprocessors, 8-19 calibration processing
A background images, 15-19 overview, 9-12
about this manual, ix background processor setup dialog, camera
abs home button, C-5 15-19 delay, 7-9
absolute range, 10-34 base engine, 18-3 exposure, 7-9
acceleration (steps), C-4 BaseCorrelator, 18-19 high-frame rate, 1-1
advanced settings dialog, 10-14 BaseEngine folder, 18-16 models, 4-2
all button, 16-5 BaseGrid, 18-17 set up, 7-4
all fields, B-40 BasePeak, 18-19 setting parameters, 7-4
all to display, B-40 BasePlugin, 18-16 setup parameters, 7-5
altering individual mask, 8-8 BaseSpotMask, 18-18 camera CCD sizes, 15-10
analysis folder, 5-4 batch editor, 11-5 camera field of view, 15-11
analyze all, 9-13 batch file, 11-4 camera fields, 15-11
analyze one button, 9-12 creating, 11-4 camera identifier, A-1
analyze options dialog, 9-21 running, 11-6 camera set up, 7-4
animate speed, 16-26 batch files, 11-2 camera target calibration, 9-28
animate tracking, 16-27 bilinear peak, 10-18, 10-23 capture, 5-3
animating images, 6-16 bin width, 16-30 capture button, 9-10
animation black line, B-39 capture calibration images dialog,
starting, B-10 blue/red button, 9-23 9-11
stopping, B-10 capture images, 7-2
application C save images, 7-2
macro, 11-1 calibration save settings, 7-2
application screen, 9-37 calibration points, 9-16 setting parameters, 7-1
apply frames, 8-4 calibration points (CPT) file, 9-25 capture tab, 7-1
apply numeric change, 8-10 calibration points files, 9-29 capture timing setup, 7-8
apply radius, 16-27 creating file from CPT files, 9-27 captured calibration images, 7-14
applying grid, 6-5 fiducial mark coordinates, 9-16 capturing images, 7-1
arithmetic operation, 8-13 fiducial marker, 9-26 cascade, 3-3
arrange icons, 3-3 four-plane fiducial marker, 9-20 change color, 6-7
arrow option, 16-8 global sizing velocimetry, 13-1 changing shape and size of particles,
arrow options, B-4 image analysis program, 9-14 16-22
arrow/contour plot, 16-6 images. (see also images, CHC choice code statistics dialog,
ASCII text files, 16-4 calibration) 16-20
auto binary conversion, 16-4 marker images, 9-14 CHC choice codes, A-4
auto open vector files check boxes, overview, 9-4 CHC statistics button, 16-19
16-24 recording images, 9-10 choice codes, A-4, A-11, A-12, A-13
Index-1
choosing color schemes, 9-23 density, 1-1 enable distributed processing, 17-3,
collapse tree, 5-5 desktop, 3-1 17-6
color dropdown, 16-27 dewarped image, 9-1, 9-32 enable software trigger, 7-12
color line, B-39 dichroic mirror, 15-6 enable time analysis, 16-25, 16-26
color options, 16-10 dichroic wavelength vs. transmission enabling distributed processing, 17-1
screen, 16-10 graph, 15-6, 15-7, 15-8, 15-9 energy content
color scheme digital gain, 7-6 choosing, B-33
choosing, 9-23 digits, C-4 energy content controls, B-34
selecting, 9-23 direct correlation, 10-18 ensemble average correlation, 9-33
component setup, 3-2 direct correlator, 10-23, 18-5 ensemble averaged mode, B-5
compute average field, 16-15 direct/snapshot, B-32 ensemble averaged velocity field, B-9
computer requirements, 2-2 disk drive, 2-2 ensemble correlation processing,
continuous, 7-11 disparity vector field, 9-33, 9-44 15-15
continuous capture mode, 7-11 disparity vector field after ensemble processing, 9-33
contour properties, 16-10 automapping, 9-46 vector field, 15-16
contour property, 16-11 display only matched particles, 13-9 ensemble vector field
contour quantity mode, B-4, B-36 display panel, 3-4 nine image pairs, 15-23
contour type, 16-12, B-5, B-39 features, 6-1 one image pair, 15-22
defining, B-6 display panel features, 6-2 epi-fluorescence, 15-2
control panel, 3-4 display while processing, 6-3 Episcopic, 15-2
correlation engine, 18-4 displaying image histogram, 6-11 error vectors, A-4
parameters, 10-23 displaying image information, 6-12 example code, 18-3, 18-4
correlation of correct velocity displaying image statistics, 6-9 exciter, 15-5
measurements, 10-4 displaying vector histogram, 6-14 exit, 3-2
correlation peak elongation, 9-32 displaying vector statistics, 6-13 expand tree, 5-5
CorrelationEngine, 10-17, 10-25, distributed processing, 4-1, 17-1 expanded uncertainty, 16-11
10-26 setting up computer, 17-1 experiment, 15-12
create image list, 6-4 setting up worker computer, 17-5 adjusting timing parameters, 7-10
create tracking, 16-27 distribution window, B-17 creating, 3-2
create vec file by vector interpolation, dock left, 4-3 creating new, 5-1
10-30 dock right, 4-3 deleting folder, 5-5
creating a run, 5-2 double correlation, 10-14, 10-15, edit, 5-2
creating an experiment, 5-1 10-24 editing, 3-2
creating calibration file, 9-27, 9-30 double correlation analysis, 10-24 file menu, 3-2
creating data files, 5-1 draw masks on images during file name components, A-1
creating dynamic mask, 8-5 processing, 8-6 how to display, 3-4
creating experiment files, 5-1 dual monitors, 4-3 managing data files, 5-2
creating experiment folder, 5-1 DVD drive, 2-2 menu options, 3-2
creating movie, B-10 dynamic mask new, 3-2
creating new calibration file, 9-12 creating, 8-5 opening, 3-2
creating new perspective calibration, draw masks on images during renaming, 5-4
9-12 processing, 8-6 setting as current, 5-4
creating run folder, 5-1 set mask velocity, 8-7 sorting order for data files, 5-3
crosshairs, 6-5 setup dialog, 8-5 time-resolved, 1-1
cumulative, 16-31 use existing, 8-6 timing setup, 7-8
dynamic processing mask, 8-5 experiment folder, 4-1, 16-24
D altering individual mask, 8-8 specifying, 4-1
data analysis, 14-6 use existing masks, 8-8 experiment tree, 5-2, 5-5, 10-13
horizontal spray, 14-6 accessing, 5-2
intensity profile, 14-6 E image list creation, 6-4, 6-13, 8-4,
statistics, 14-6 each field, B-40 8-6, 8-10, 8-13–8-17, 9-37, 9-38,
vertical spray, 14-6 edge pixels, 9-22 9-40, 9-43, 9-45, 10-12, 10-14,
data files, 18-27, A-1 edit camera, 9-13 10-37, 11-2, 11-5, 12-4, 12-7,
date-time code, 11-1 edit experiment, 3-2 12-8–12-12, 13-2, 13-3, 15-16–
define target button, 9-12 edit folders dialog, 16-24 15-23, 16-2–16-4, 16-6, 16-8,
defining contours, B-6 edit run, 3-2, 3-3 16-10, 16-11, 16-15, 16-18–
deformation grid, 10-21 Edit Run tool button, 3-3 16-33, 17-1–17-8, B-37, C-1–C-5
Delaunay Triangulation, 10-30 editing experiment, 5-2 tab, 5-2
delay, 5-3 editing run, 5-2 experiments11 properties
delete experiment, 5-5 ellipse, 8-5 sharing tab, 17-2
delete run, 5-5 emission filter, 15-8 export, 16-26
delete tracking, 16-27 export images dialog, 6-15
Index-2 Insight 4G Global Image Acquisition, Analysis and Display Software

export to figure, B-19 global sizing velocimetry (continued) image binning, 8-12
exporting image statistics, 6-10 processing images, 13-2 setup, 8-14
exporting images, 6-15 sizing processor, 13-3 image calculator, 8-12, 8-18
exposure, 7-5, 7-9 tracking processor, 13-6 setup, 8-13
free mode, 7-11 view data file on top of raw image, setup dialog for background image
mode, 7-8 13-8 subtraction, 8-19
options, 7-11 viewing processing results, 13-8 setup for two preprocessor
synchronized mode, 7-11 global velocity diagnostics, 1-1 background image subtraction,
external trigger, 7-4, 7-10 grayscale invertion, 8-13 8-20
eyepiece field of view, 15-11 grid, 9-5 image capture settings, 7-15
grid connection order, 9-22 image capture timing (*.TXT) file
F grid engine, 10-13, 18-4 format, A-15
FFT correlation, 10-17 grid engine parameter, 10-20, 10-23 image conditioning, 8-7
FFT correlator, 10-23 grid option, 6-5 image deformation, 8-12, 8-16
fiducial mark coordinates, 9-16 grid tolerance pixels, 9-22 image interpolation, 8-16
fiducial marker, 9-12, 9-26 GridEngine, 10-15 image shift, 8-16
field of view, 9-2, 15-10 GSV. (see also global sizing image dewarping, 8-12
file extensions, A-2 velocimetry) setup, 8-15
file sharing, 17-3 calibration, 13-1 image filter, 8-12
file type, 5-3 change display settings for particle setup, 8-14
filename components, A-1 size and/or velocity, 13-9 image filtering, 8-7
filiter cube, 15-1 open data in a table, 13-8 local median, 8-7
filling holes, 10-35 particle setup, 13-9 local median subtraction, 8-7
final validation tolerance, 10-30 process monitor, 13-7 none, 8-7
flashlamp, 7-14 processing images, 8-1, 13-2 image generator, 8-12, 8-17
flashlamp frequency, 7-4 processing results, 13-8 average intensity, 8-18
flip image, 7-6, 8-13 sizing processor, 13-3 generator output setup, 8-18
flood, B-39 tracking processor, 13-6 minimum intensity, 8-18
flood + line, B-39 view data file on top of raw image, setup, 8-15
flow properties, 16-11, 16-12 13-8 image histogram, 6-11
flow velocity, 9-1 GSV application, 8-1 image information, 6-12
fluctuations mode, B-5 GSV process monitor screen, 6-12
fluorescence, 1-2, 15-2 graph panels, 13-7 image interpolation, 8-16
fluorescent emission light, 15-8 image masking, 8-7
fluorescent particles, 15-4 H invert, 8-7
four-plane fiducial marker, 9-20 hardware and software requirements, mask padding, 8-7
four-plane target, 9-11 2-2 thresholding, 8-7
frame grabber, 7-6, 7-14 hardware compatibility chart, 4-2 view, 8-7
models, 4-2 hardware component parameter image preprocessing, 8-11, 10-28
frame rate, 7-5 setup, 7-2 image preprocessors, 8-12
frame straddle frame identifier, A-2 hardware components image processing, 8-2
free, 7-11 specifying, 4-2 image shift, 8-16
hardware setup, 3-2 image size pixels, 9-22
G dialog box, 4-2 image statistics, 6-9
Gaussian diameter, 10-22 help Image Tools option, 3-2
Gaussian filter, 8-14 menu options, 3-3 images
Gaussian mask, 10-22, 18-5 histogram, 9-23, 9-24 animation, 6-16
Gaussian peak, 10-18, 18-5 displaying, 9-24 calibration, 7-14, 9-10, 9-11, 9-17
Gaussian spot mask engine, 18-11 histogram plot, 16-8, 16-28 calibration marker, 9-14
generate, 8-4 histogram plot for vector and particle calibration result, 9-26
generator output setup, 8-18 files, 16-8 capture option, 3-4
getting started, 2-1 histogram plot with cumulator capture settings, 7-15
global fluorescence, 1-2 enabled, 16-29 capturing, 7-1
global image capture, 1-1 horizontal scan interval, 13-4 capturing for calibration, 3-2
global imaging, 1-1 host experiments directory button, capturing images, 7-15
global measurement, 1-1 17-7 continuous capture mode, 7-11
global sizing velocimetry, 8-1, 13-1 create image list, 6-4
calibration, 13-1 I deleting capture settings, 7-16
change display settings for particle I/O board, 4-3 display while processing, 6-3
size and/or velocity, 13-9 ICorrelationMap, 18-21 displaying
open data in a table, 13-8 IGridPoint, 18-22 while capturing, 7-12
process monitor, 13-7 displaying using Tecplot, 3-3
User’s Guide Index-3

images (continued) laser mask padding, 8-7
exporting, 6-15 laser pulse delay, 7-9 masking operation, 8-13
exposure mode options, 7-11 light sheet, 9-9 MATLAB, 3-2
frame straddle mode, 7-9 Nd:YAG, 7-13 MATLAB plugin setting, 12-6
histogram,, 6-11 systems, 7-9 MATLAB software, 1-1, 10-1, B-1
information, 6-8 off button, 7-14 animating displays, B-10
loading capture settings, 7-16 on button, 7-14 creating movies, B-10
LUT, 6-6 PLIF laser pulses, 7-9 profile analysis, B-16
PLIF exporting, 12-12 pulse energy, 7-13, 7-14 proper orthogonal decomposition
processing, 8-1, 10-1 pulse repetition rate, 7-9 toolbox, B-30
recording calibration, 9-10 pulse separation value, 7-9 animation option, B-41
saving pulses per trigger value, 7-10 contour types, B-39
capture settings, 7-15 Q-Switch, 7-9, 7-13, 7-14 energy content, B-33
captures, 7-16 setup, 7-3 movie option, B-41
to disk, 7-12 parameters, 7-4 multi-mode, B-37
to RAM, 7-12 starting, 7-14 options, B-31
selecting, 6-5 stopping, 3-3 reconstruction of fields, B-38
sequence capture mode, 7-11 turning off, 7-14 single POD modes, B-37
single capture mode, 7-11 warning, 15-1, 15-13 spatial toolbox, B-3
single frame mode, 7-9 YAG, 7-9 navigating, B-4
specifying for PLIF and PIV, 7-9 laser camera arrangement, 13-3 options, B-4
spray processing, 14-1 laser frequency, 7-4 starting, B-1
statistics, 6-9 laser off, 3-3 time series analysis
zooming in and out, 6-3 laser pulse delay, 7-9 performing, B-21
improperly identified markers laser set up, 7-3 using time series analysis toolbox,
fixing, 9-25 lasers, high-pulse rate, 1-1 B-11
installation, 2-2 light sheet, 9-9, 15-2 max distance to nearest particle,
DVD, 2-1 light sheet generator, 14-1 10-30
instantaneous mode, B-4 line profile, 6-14 maximum deviation from local flow
instantaneous velocity field, B-7, B-8 linear fit method, 12-5 direction, 13-6
Intel Core i3 computer, 2-2 local flow calculation, 10-26 maximum distance between neighbor
intensity reduction, 8-7 local flow estimation, 10-29 particles, 13-6
intensity restoration, 8-7 local mean, 10-33 maximum in-plane displacements,
interface, 18-3 filter, 8-14 10-6
interface classes, 18-20 local median, 10-33 maximum intensity image, 15-20
interface IGridPoint.h, 18-21 filter, 8-14 maximum normalized width, 13-5
interrogation errors, 10-5 local range filter, 8-15 maximum number of neighbors, 13-6
interrogation spot size, 10-3 log/normal scale button, 9-25 maximum oscillation contrast, 13-5
interrogation spot with one particle- Look Up Table. (see also LUT) maximum oscillation overlap, 13-5
image diameter displacement, 10-9 lost pairs, 10-6 maximum out-of-plane displacement,
introduction, 1-1 LUT, 6-6, 9-24 10-7
IPluginData, 18-22 option, 6-6, 6-7 maximum size difference, 13-6, 13-7
IPluginParameter, 18-23 mean and standard deviation (*.STD)
IPluginSettingFactory2, 18-24 M vector file format, A-9
IPluginVector, 18-24 macro mean and standard deviation *.STD
IPluginVectorField2, 18-25 batch editor, 11-5 file listing, A-10
IProcessGrid, 18-26 creating, 11-2 measurements
IProcessingMask, 18-26 creating batch file, 11-4 global spray diagnostics, 1-2
ISpotImage, 18-26 running, 3-4 time-resolved, 1-1
stopping, 3-4 menu options, 3-2
J macro P#, 11-1 message folder, 16-24
jog, C-5 macros, 11-1, 11-2 message option, 16-23
batch files, 11-2 message page, 16-23
K magnification, 15-10 micro particle image velocimetry, 15-1
key, 2-1 managing data files, 5-1, 5-2 microflows, 1-1
managing experiment files, 5-1 microPIV
L manual, B-40 example experiment, 15-12
Lambda Cl, 16-11 manual (calibration number), 9-3 experiment, 15-12
Laplacian filter, 8-14 manual move mode, C-1, C-3 experiment kit, 15-12
Laplacian of Gaussian filter, 8-14 map network drive, 17-5 experiment setup, 15-12, 15-13
mark maxima/clear marker, 16-27 filter cube, 15-1
mask, 8-3 low speed laser, 15-2

microPIV (continued) optical microscopy, 15-2 PIV (continued)
operating principles, 15-4 Optical Microscopy, 15-2 processing images, 8-1
overview, 15-2 options button, 9-12 processor setup box, 10-13
microPIV system, 1-1 oscillation intensity threshold, 13-3 setting up processing parameters,
microscopy web sites, 15-2 output options, B-37 10-13
Microsoft overlay/hide tracking, 16-27 specifying settings, 3-4
Windows 7 professional, 2-2 overview, 1-1, 15-2 PIV processor, 10-1
minimum and maximum horizontal overwrite timestamp, 16-25, 16-26 setup, 9-41, 10-13
shift, 13-6 PIV processor and PIV advanced
minimum and maximum vertical shift, P settings windows, 10-38
13-6 package contents, 2-1 PIV uncertainty
minimum in-plane displacement, 10-8 parameters component-wise uncertainties,
minimum intensity, 8-18 setting camera, 7-4 10-37
minimum intensity image, 15-21 setting timing, 7-8 exporting, 10-38
minimum normalized width, 13-5 particle (*.PAR) files, A-11 method, 10-36
minimum oscillation visibility, 13-5 choice codes, A-11 peak ratio, 10-36
minimum power spectrum peak ratio, velocity not set, A-11 pixel *.VEC file listing, A-5
13-5 particle color, 13-9 pixel *.VEC files, A-4
mode, C-5 particle file pixel *.VEC header line, A-5
model, 7-4 arrow/contour plot, 16-6 pixel aspect ratio, 7-5
modified vectors, 16-10 dialog box, 16-3 pixel displacement, 7-9, 9-1
Molecular Expressions, 15-2 histogram plot, 16-8 pixel intensity, 7-6
monitor, 2-2 scatter plot, 16-7 distribution, 9-24
move, C-5 particle group size, 10-30 pixels, 7-5, 9-3, 9-14, 9-22, 9-23
movie particle identification, 10-26, 10-28 planar laser induced fluorescence, 8-1
animation, B-41 particle minimum size, 10-28 PLIF, 1-2, 12-1
creating, B-10 particle peak intensity threshold, application, 8-1
options, B-41 10-28 calibrating and processing, 12-1
multi-camera PLIF, A-13 particle scale, 13-9 camera delay, 7-9
multi-mode, B-37 particle settings, 13-9 camera exposure, 7-9
output, B-41 particle style, 13-9 capture timing setup, 14-3
multiplier, C-4 particle tracking, 10-27, 10-30 capturing calibration images, 12-1
multi-thread processing, 10-15 particle validation tolerance, 10-30 exporting data, 12-12
peak engine, 18-4 laser pulses, 7-9
N peak ratio, 10-36 linear fit method, 12-5
Nd:YAG laser, 7-13, 15-5 peak to noise peak ratio, 10-36 LUT table, 6-6
Nd:YAG systems, 7-9, 7-14 PeakEngine, 10-17, 10-18 MATLAB plugin, 12-6
negative/positive peak-to-noise ratio, 10-23 using, 12-7
setting, 9-22 perspective calibration, 8-2, 9-1, 9-4, measurements, 1-1
new experiment, 3-2 9-24, 9-26, 9-31 processing images, 8-1, 12-2
screen, 5-1 process, 9-4 processing method, 12-5
new project information, 18-7 photobleaching, 15-5 processor setup, 12-3
new run, 3-2, 3-3 PIV, 10-36 ratiometric method, 12-5
screen, 5-2 2-D calibration, 9-1 scalar data export, 12-12
Nikon Microscopy U, 15-2 application, 8-1 setting up processing parameters,
no mask, 10-17, 10-22 calibrating, processing, viewing, 12-3
node average, 16-16 10-1 specifying settings, 3-4
number of bins, 16-30 calibration process PLIF processing
number of images with same field of view, 9-2 two camera, 12-8
sequence number, A-2 manual (calibration numbers), 9-3 plot options field, B-35
number of stream traces, 16-20 object measure, 9-3 plot type option, 16-6
Nyquist grid, 10-16, 10-20 camera exposure, 7-9 plugin
capture timing setup, 14-2 coordinates and units, 18-3
O cross-correction processing, 10-11 data, 18-2
object measure, 9-3 frame model, 7-9 definition, 18-1
object search area pixels, 9-22 images to capture, 7-9 development kit, 18-3
object size, 9-21 plugin engine settings, 10-19 DLLs, 18-1
object size pixels, 9-22 plugin engines, 10-15 parameter, 18-2
Olympus Microscopy Resource processing, 10-1 SDK files, 18-16
Center, 15-2 algorithms, 10-11 structure, 18-1
once-per-revolution mechanism, 7-10 images, 10-1 plugin data, 10-19, 18-2
open experiment screen, 5-2 images after capture, 7-15 plugin parameter, 10-19

plugin wizard, 18-3, 18-5 PTV scalar data export, 12-12
generated code, 18-4 processor, 10-26 scalar date (*.SCA) files, A-11
PluginData, 10-26 processor setup box, 10-13 multi-camera PLIF, A-13
PluginID, 18-27 pulse energy, 7-13, 7-14 simultaneous PLIF-PIV, A-12
PluginParameter, 18-5 pulse repetition rate, 7-9 single-camera PLIF, A-12
polarity, 7-3 pulse separation value, 7-9 scalar file
polygon, 8-5 pulsed spray images, 14-4 dialog box, 16-4, 16-5, 16-11
positive/negative setting, 9-22 scalar property, 1-2
post trigger capture, 7-12 Q scale, 6-8
post-processing pipeline Q-Switch, 7-9, 7-13, 7-14 scatter plot, 16-7, 16-32, 16-33
macro, 11-1 scatter plot for vector and particle
PowerView R files, 16-7
camera CCD sizes, 15-10 rake position, 16-21 Scheimpflüg angle, 9-8, 9-9
PowerView 4MP-HS camera, 15-10 RAM, 2-2 screen updating, 6-8
preplot path name, 16-24 range, 16-30 SDK, 18-5
preprocessing pipeline, 8-11 ratiometric method, 12-5 select all, B-32
applying, 8-12 raw data folder, 5-4 select CPT files, 9-13
macro, 11-1 reconstruction of fields, B-38 select laser power level, 7-14
preprocessor pipeline editor, recording calibration images, 9-10 selecting color scheme, 9-23
generated output setup, 8-18 rectangle, 8-5 selecting columns, B-14
printing image statistics, 6-10 rectangular grid, 10-16, 10-20, 18-5 selecting components, 4-1
process, 5-3 recursive ensemble, 15-18 selecting entire map, B-15
process monitor, 10-25 recursive filling, 10-35 selecting options for analysis, B-12
processing recursive Nyquist grid, 10-16, 10-20 selecting points, B-12
macro, 11-1 refresh tree, 3-2 selecting regions, B-13
processing and data post-processing, region of interest, 8-2, 8-9, 9-21 selecting rows, B-14
8-3 applying, 8-10 send all, 7-3
processing images, 8-1 deleting, 8-10 sending image statistics table to
processing mask, 8-2, 8-3, 8-5, 8-9 related product literature, ix clipboard, 6-10
defining static processing mask, relationship of eyepiece image area, sequence, 7-12
8-3 15-11 sequence capture mode, 7-11
deleting, 8-9 relative refractive index of particles, service policy, iv
macro, 11-1 13-4 set experiment as current, 5-4
processing progress, 8-3 remember settings, 6-7 set mask velocity, 8-5, 8-7
processing tab, 8-1 rename experiment, 5-4 set mean button, 16-17
application, 8-1 rename run, 5-4 set run as current, 5-4
image preprocessing, 8-2 residual pixels, A-16 set threshold, 16-27
processing and data post- resolution, 7-5 SetRe10 button, C-5
processing, 8-3 reverse order, 8-13 setting
processing mask, 8-2 Reynolds stress, 16-15 high, 7-14
region of interest, 8-2 ROI, 8-9 low, 7-14
spatial calibration, 8-2 rotating machinery, 7-10 medium, 7-14
viewing processing progress, 8-3 rotation, 8-13 off, 7-14
processor rotation angle, 7-10 setting output lookup table gain and
setting up, 10-28 rules-of-thumb for PIV analysis, 10-2 pseudocolor, 15-15
processor pipeline editor, 8-11, 9-39, run setting up software, 4-1
9-42, 10-31, 10-32 creating, 3-2 setting up vector validation, 10-31
profile analysis, B-16 creating new, 3-3, 5-2 setting up vectors, 10-30
programming macros, 11-1 deleting folder, 5-5 setup, 3-2
batch files, 11-4 editing, 3-3, 5-2 synchronizer, 7-3
working with, 11-2 menu options, 3-2 setup batch, 17-8
proper orthogonal decomposition new, 3-2, 3-3 Setup INSIGHT dialog box, 4-1
toolbox, B-30 opening, 3-2 setup screen, 17-4, 17-6, 17-7
animation option, B-41 renaming, 5-4 show calibration points, 9-31
contour type, B-39 setting as current, 5-4 show delay, 5-4
energy content, B-33 show grid lines, 9-31
movie option, B-41 S show hardware, 5-4
multi-mode, B-37 saturation of colors, 6-7 show process, 5-4
options, B-31 save button, 16-25 show text, 13-10
reconstruction of fields, B-38 save turb, 16-15 show traverse, 5-4
single POD modes, B-37 scalar contour show vectors, 13-10
pseudocolor, 6-7, 9-24 minimum and maximum, 16-14 simultaneous PLIF-PIV, A-12

single, 7-11 statistics options, 16-19 experiment folder, 16-24
single capture mode, 7-11 statistics table, 6-9 file option, 16-2
single modes velocity field, B-39 status bar, 3-4 histogram plot, 16-28
single POD modes, B-37 STD file header, A-10 installation, 16-1
single-camera PLIF, A-12 stereo automapping, 8-1, 9-1, 9-32 message folder, 16-24
single-plane fiducial marker, 9-16 procedure, 9-36 message option, 16-23
single-plane target, 9-10 stereo PIV application, 8-1 number of bins, 16-30
size shape analysis, 8-1, 8-21 stereo PIV camera number of stream traces, 16-20
setting up, 8-21 location, 9-7 opening, 16-2
setting up sizing processor, 8-22 STL file header, A-14 preplot path name, 16-24
Size Shape Analysis, 8-1 stopping animation, B-10 rake position, 16-21
smoothing, 10-36 stream option, 16-20 range, 16-30
SNR vector validation, 18-2 stream trace, 16-20 reading files, 16-2
software, 3-3 sub-integer histogram plot, 16-31 save button, 16-25
acquisition, 1-1 sub-integer plot, 16-31 scatter plot, 16-32
features, 3-1 submitting comments, ix selecting files, 16-2
license, iv subtract mean velocity, 16-18 selecting modified vectors, 16-10
presentation, 1-1 subtract source, 8-15 selecting plot types, 16-6
processing, 1-1 subtract vec, 16-19 selecting velocity vectors, 16-10
sort order screen, 5-3 super-resolution particle velocimetry, statistics options, 16-19
sort tree, 3-2, 3-3 10-26 stream option, 16-20
option, 3-3 application, 8-1 sub-integer plot, 16-31
spatial calibration processor flowchart, 10-27 subtract mean velocity, 16-18
macro, 11-1 processor setup, 10-28 subtract option, 16-18
spatial toolbox options, B-3, B-4 supported cameras, 4-2 subtract reference field button,
specifying experiment folder, 4-1 models, 4-2 16-18
specifying hardware components, 4-2 supported frame grabbers, 4-2 subtract vec, 16-19
spiral toolbox window, B-1 models, 4-2 table of flow properties for scalar
spot mask engine, 10-17, 18-4 synchronized, 7-11 fields, 16-12
parameters, 10-22 synchronizer, 7-9, 7-14 table of flow properties for vector
spot mask wizard, 18-6 external trigger, 7-10 fields, 16-11
SpotMask, 10-25, 10-26 setup, 7-3 troubleshooting, 16-34
SpotMaskEngine, 10-15 trigger timeout value, 7-10 turbulence intensity, 16-16
spray, 1-1 vector field arrow color, 16-13
data analysis, 14-6 T vector files, 16-2
hardware setup, 14-1 table of contour types, 16-12 velocity/modified velocity check,
processing images, 14-1 target 16-19
software setup for pulsed spray, defining, 9-15 view option, 16-21
14-4 defining, 9-12 viewing statistics of particles, 16-23
software setup for steady spray, technical papers, 15-3 y-axis parameter selection, 16-29
14-2 TecPIV, 16-2 temperature, 1-1
specifying settings, 3-4 Tecplot, 16-1 test file access, 17-4
SSA Tecplot software, 1-1, 3-3, 10-1, 16-1, test image, 7-6
application, 8-1 16-34 text color, 13-10
dialog box, 8-21 arrow option, 16-8 text font, 13-10
SSA (size shape analysis, 8-21 auto open vector files check boxes, text settings, 13-10
standard deviation, 16-16 16-24 this is a host PC button, 17-3
range, 10-34 average fields, 16-15 threshold assistant, 8-23, 10-29
standard uncertainty, 10-36, 16-11 average option, 16-15 threshold assistant apply, 8-23
start draw button, 8-5 bin width, 16-30 threshold level, 9-15
starting, 2-2 changing shape and size of threshold value, 9-23
starting animation, B-10 particles, 16-22 setting, 9-12, 9-14, 9-23
stat option, 16-23 CHC choice code statistics dialog, tile horizontal, 3-3
static mask 16-20 tile vertical, 3-3
set mask velocity, 8-5 CHC statistics button, 16-19 time analysis probe, 16-26
static tab, 8-4 clear, 16-20 time out, C-4
statis mask color option, 16-10 time series analysis, B-21
generate, 8-4 contact information, 16-35 time step skip, 16-26
statis processing mask, 8-3 contour properties, 16-10 time tab, 16-25
statistics, 6-9 contour type, 16-12 time toolbar, B-11
statistics file list *.STL file listing, A-14 cumulative, 16-31 timeout, 7-6
statistics list (*STL) files, A-13 edit folders dialog, 16-24 time-resolved measurements, 1-1

time-resolved toolbox, 16-25 vector color, 13-10 Windows, 1-1
time, 16-25 vector conditioning, 10-35 world location, 9-32
tracking, 16-26 vector field arrow color, 16-13
timing vector field conditioning setup, 10-35 X
parameters, 7-10 vector field display, 16-9 x, y, z & value, 16-27
viewing diagram, 7-10 vector file
timing diagram for PLIF mode arrow/contour plot, 16-6 Y
acquisition, 14-5 dialog box, 16-3, 16-10 YAG laser, 7-9
timing file listing, A-15, A-16 histogram plot, 16-8 y-axis parameter selection, 16-29
timing setup, 7-8 scatter plot, 16-7 YLF laser, 15-1
parameters, 7-9 vector file format, A-3
title, C-4 vector global validation, 10-34 Z
tool bar options, 3-3 setup, 10-34 z polynomials, 9-5
tools crosshair option, 9-8 vector histogram, 6-14 ZeroPad mask, 10-17
tools menu options, 3-2 vector information, 6-13 ZeroPadMask, 10-22
track absolute value, 16-27 vector local validation, 10-33 zoom in, 9-9
track multiple vortices, 16-27 setup, 10-33
tracking dialog box in focusPIV, 16-27 vector plot for vector and particle files,
tracking tab, 16-25, 16-27 16-6
traverse, 5-3, 9-10 vector scale, 13-10
enabling and setting, 7-8 vector settings, 6-8, 13-10
enabling and setting, C-2 vector statistical table, 6-13
using, C-1 vector validation
traverse manager setting up, 10-31
button, C-1 vector/scatter graph, 16-22
control, C-3 vector/scatter plot for vector and
move mode, C-1 particle files, 16-7
screen, C-2 vectors, 1-1, 6-8
using, C-5 setup, 6-8
traverse setup, C-2 validation, 10-30
trigger delay, 7-10 velocity, 1-1, 7-5, 7-9, 9-1
trigger timeout, 7-10 measurement, 10-30
troubleshooting, 16-34 peak, 10-30
TSI Tecplot Add-On software, 16-1 simultaneous measurement, 1-2
TTL signal, 7-3 vector, 9-1
turbo mode, 6-3 velocity (steps), C-4
turbulent mode, B-5 velocity *.VEC file, A-6
two camera PLIF processing, 12-8 velocity flow, 9-1
two-plane fiducial marker, 9-18 velocity gradient error, 10-3
two-plane target, 9-11 velocity grid interval, 10-30
typical image velocity interpolation, 10-27, 10-30
with minimum intensity subtracted, velocity magnitude, 16-11
15-22 velocity not set, A-11
velocity vectors, 16-10
U velocity/modified velocity check, 16-19
uncertainty on a number line, 10-36 version, 7-3
units, C-4 vertical scan output, 13-4
universal median, 10-33 vertical scan window height, 13-4
unpacking and checking, 2-1 vertical scan window width, 13-4
use existing mask, 8-6 video card, 2-2
use existing masks, 8-8 view linked CPT files, 9-31
use measured vectors only, 10-35 view option, 16-21
use motion match, 13-6 view page, 16-21
using dual monitors, 4-3 screen, 16-21
viewing statistics of particles, 16-23
V
validating vectors, 10-30 W
vector (*.VEC) file, A-3 warning
choice codes, A-4 laser, 15-1, 15-13
error vectors, A-4 warranty, iii
header line, A-7 white light illumination, 15-9
listing, A-7 window menu options, 3-3

Reader’s Comments
Please help us improve our manuals by completing and returning this questionnaire to the address listed in the
“About This Manual” section. Feel free to attach a separate sheet of comments.
Manual Title Insight 4G Global Image Acquisition, Analysis and Display Software P/N 6004904
1. Was the manual easy to understand and use?

Yes No
Please identify any problem area(s) _________________________________________________________

_____________________________________________________________________________________
_____________________________________________________________________________________
2. Was there any incorrect or missing information? (please explain) __________________________________
_____________________________________________________________________________________
_____________________________________________________________________________________
3. Please rate the manual according to the following features:
Good Adequate Poor
Readability
Accuracy
Completeness (is everything there?)
Organization (finding what you need)
Quality and number of illustrations
Quality and number of examples
Comments: ____________________________________________________________________________
_____________________________________________________________________________________
_____________________________________________________________________________________
4. Which part(s) of this manual did you find most helpful? __________________________________________
_____________________________________________________________________________________
_____________________________________________________________________________________
5. Rate your level of experience with the product:
Beginning Intermediate Expert
6. Please provide us with the following information:
Name _________________________________ Address______________________________________
Title __________________________________ ______________________________________
Company ______________________________ ______________________________________

TSI Incorporated – Visit our website www.tsi.com for more information.
USA Tel: +1 800 874 2811 India Tel: +91 80 67877200

UK Tel: +44 149 4 459200 China Tel: +86 10 8219 7688
France Tel: +33 1 41 19 21 99 Singapore Tel: +65 6595 6388
Germany Tel: +49 241 523030
P/N 6004904 Rev. C (3/19/2015) ©2015 TSI Incorporated Printed in U.S.A.

INSIGHT4G Software Manual

Uploaded by

Copyright:

Available Formats

INSIGHT4G Software Manual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

INSIGHT4G Software Manual

Uploaded by

Copyright:

Available Formats

What is the purpose of the manual?

What is the purpose of the manual?

What is the warranty period for the product?

What is the warranty period for the product?

INSIGHT 4G™

P/N 6004904, REVISION C

This is a manual history of the Insight 4G™ software manual

This manual is used with Insight 4G (all versions) software.

Part Number 6004904 / Revision C / March 2015

iv Insight 4G Global Image Acquisition, Analysis and Display Software

Manual History .......................................................................................... ii

vi Insight 4G Global Image Acquisition, Analysis and Display Software

viii Insight 4G Global Image Acquisition, Analysis and Display Software

Related Product Literature

x Insight 4G Global Image Acquisition, Analysis and Display Software

Insight 4G™ software is a Microsoft® Windows®-based software that is

Insight 4G software is a 64-bit Windows® 7 application.

Insight 4G software acquires, analyzes global images (velocity field, or

Insight 4G software automates the process of image capture, analysis and

Insight 4G software is functionally divided into two parts—an acquisition

Insight 4G software is a true Global Image capture, analysis, and display

1-2 Insight 4G Global Image Acquisition, Analysis and Display Software

Be sure to do the following before starting:

Unpacking and Checking Package Contents

Note: The proper installation of Insight software is determined by the

Installing Insight Software

Starting Insight Software

2-2 Insight 4G Global Image Acquisition, Analysis and Display Software

The Insight Desktop

Refer to Chapter 6, “Viewing, Enhancing, and Displaying Image Files,” for

3-2 Insight 4G Global Image Acquisition, Analysis and Display Software

Insight Tool Bar

Getting Acquainted with Insight Software 3-3

3-4 Insight 4G Global Image Acquisition, Analysis and Display Software

The following sections provide details on how to perform these tasks.

Specifying the Experiment Folder

Note: To avoid causing errors in Insight software, it is recommended that

To Specify an Experiment Folder

To Specify the Hardware Components in your

b. The list of cameras will automatically reduce to those supported by

4-2 Insight 4G Global Image Acquisition, Analysis and Display Software

Using Dual Monitors

To Use Two Monitors

Setting Up Insight Software 4-3

4-4 Insight 4G Global Image Acquisition, Analysis and Display Software

Insight™ software’s Experiment/Run menu options and the Experiment

Creating Experiment and Run Folders

Editing an Experiment or Run

The Experiment Information or Run Information dialog box appears.

Managing Insight Files

Accessing the Experiment Tree

5-2 Insight 4G Global Image Acquisition, Analysis and Display Software

Specifying the Sorting Order for Data Files

To Specify the Display Order of Data Files

Creating and Managing Experiment and Data Files 5-3

Setting an Experiment or Run as Current

To Specify an Experiment or Run as Current

Renaming an Experiment/Run Folder

5-4 Insight 4G Global Image Acquisition, Analysis and Display Software

Deleting an Experiment/Run Folder