IMAGE PROCESSING

Abstract: In this paper, the basics of image processing is found in the photographic
capturing an image, image processing to dark room
modify and enhance the image are discussed. Electronic image processing is further
There are many applications for Image classified as:
Processing like surveillance, navigation, and (i). Analog processing
robotics. Robotics is a very interesting field (ii). Digital processing
and promises future development so it is
chosen as an example to explain the various ANALOG PROCESSING:
aspects involved in Image Processing. Analog processing is the control of
The various techniques of Image brightness and contrast of television
Processing are explained briefly and the image. The television signal is a voltage
advantages and disadvantages are listed. There level that varies In amplitude to represent
are countless different routines that can be brightness through out the image by
used for variety of purposes. Most of these electrically altering these signals , we
routines are created for specific operations and
correspondingly alter the final displayed
applications. However, certain fundamental
techniques such as convolution masks can be image appearance.
applied to many classes of routines. We have
concentrated on these techniques, which DIGITAL IMAGE PROCESSING:
enable us to adapt, develop, and use other Processing of digital images by means
routines and techniques for other applications. of digital computer refers to digital image
The advances in technology have created processing. Digital images are composed of
tremendous opportunities for visual system finite number of elements of which has a
and image processing. There is no doubt that particular location value. Picture elements,
the trend will continue into the future. image elements, and pixels are used as
elements for digital image processing.
Digital Image Processing is concerned
1. INTRODUCTION with processing of an image. In simple words
an image is a representation of a real scene,
IMAGE PROCESSING: either in black and white or in color, and either
Image processing pertains to the in print form or in a digital form i.e.,
alteration and analysis of pictorial (technically an image is a two-dimensional
information. Common case of image light intensity function. In other words it is a
processing is the adjustment of brightness and data intensity values arranged in a two
contrast controls on a television set by doing dimensional form, the required property of an
this we enhance the image until its subjective image can be extracted from processing an
appearing to us is most appealing. The image. Image is typically by stochastic
biological system (eye, brain) receives, models. It is represented by AR model.
enhances, and dissects analyzes and stores Degradation is represented by MA model.
images at enormous rates of speed. Basically Other form is orthogonal series
there are two-methods for processing pictorial expansion. Image processing system is
information. They are: typically non-casual system. Image processing
I. Optical processing is two dimensional signal processing. Due to
II. Electronic processing. linearity Property, we can operate on rows and
columns separately. Image processing is vastly
Optical processing uses an being implemented by “Vision Systems” in
arrangement of optics or lenses to carry out robotics. Robots are designed, and meant, to
the process. An important form of optical be controlled by a computer or similar
devices. While “Vision Systems” are most COMPRESSION
sophisticated sensors used in Robotics. They Reducing the storage required to save
relate the function of a robot to its an image or the bandwidth required to
environment as all other sensors do. “Vision transmit it. Ex. JPEG (Joint Photographic
Systems” may be used for a variety of
applications, including manufacturing, Experts Group) image compression standard.
navigation and surveillance. Some of the
applications of Image Processing are:
1. Robotics. MORPHOLOGICAL PROCESSING
2. Medical Field. Tools for extracting image components
3. Graphics and Animations. that are useful in the representation and
4. Satellite Imaging. description of shape.

IMAGE SEGMENTATION
2. INDEX TERMS Computer tries to separate objects
separate objects from the image background
IMAGE PROCESSING from the image background. It is one of the
Image processing is a subclass of most difficult tasks in DIP. A rugged
signal processing concerned specifically with segmentation procedure brings the process a
Pictures. Improve image quality for human long way toward successful solution of an
perception and/or computer interpretation. image problem. Output of the segmentation
Image enhancement. To bring out detail is stage is raw pixel data, constituting either the
obscured, or simply to highlight certain boundary of a region or all the points in the region
features of interest in an image. itself.

IMAGE RESTORATION
Improving the appearance of an image
tend to be based on mathematical or
probabilistic models of image degradation.

COLOR IMAGE PROCESSING

Gaining in importance because of the
3. ANALYSIS
significant increase in the use of digital images The following is the overall view and
over the Internet. analysis of Image Processing.

IMAGE PROCESSINGTECHNIQUES:
WAVELETS
Image Processing techniques are used
Foundation for representing images in
to enhance, improve, or otherwise alter an
various degrees of resolution. Used in image
image and to prepare it for image analysis.
data compression and pyramidal
Usually, during image processing information
representation (images are subdivided
is not extracted from the image. The intention
successively into smaller regions)
is to remove faults, trivial information, or
information that may be important, but not
useful, and to improve the image. Image Representation and Description
processing is divided into many sub processes, transforms raw data into a form suitable for
including Histogram Analysis, Threshholding, the Recognition processing.
Masking, Edge Detection, Segmentation, and
others. V. KNOWLEDGE BASE:
A problem domain detailing the
STAGES IN IMAGE PROCESSING regions of an image where the information of
interest is known to be located is known as
knowledge base. It helps to limit the search.

VI. THRESHOLDING:
Thresholding is the process of dividing
an image into different portions by picking a
certain grayness level as a threshold,
comparing each pixel value with the threshold,
and then assigning the pixel to the different
portions, depending on whether the pixel’s
grayness level is below the threshold or above
the threshold value. Thresholding can be
performed either at a single level or at
multiple levels, in which the image is
processed by dividing it into ” layers”, each
with a selected threshold.
Various techniques are available to
choose an appropriate threshold ranging from
simple routines for binary images to
I. IMAGE ACQUISITION: sophisticated techniques for complicated
An image is captured by a sensor (such images.
as a monochrome or color TV camera) and
digitized. If the output of the camera or sensor VII. CONNECTIVITY:
is not already in digital form, an analog-to Sometimes we need to decide whether
digital converter digitizes it. neighboring pixels are somehow “connected”
or related to each other. Connectivity
II.RECOGNITION AND INTERPRETATION: establishes whether they have the same
Recognition is the process that assigns property, such as being of the same region,
a label to an object based on the information coming from the same object, having a similar
provided by its descriptors. Interpretation is texture, etc. To establish the connectivity of
assigning meaning to an ensemble of neighboring pixels, we first have to decide
recognized objects. upon a connectivity path.

III. SEGMENTATION: VIII. NOISE REDUCTION:

Segmentation is the generic name for a Like other signal processing mediums,
number of different techniques that divide the Vision Systems contains noises. Some noises
image into segments of its constituents. The are systematic and come from dirty lenses,
purpose of segmentation is to separate the faulty electronic components, bad memory
information contained in the image into chips and low resolution. Others are random
smaller entities that can be used for other and are caused by environmental effects or
purposes. bad lighting. The net effect is a corrupted
image that needs to be preprocessed to reduce
or eliminate the noise. In addition, sometimes
IV. REPRESENTATION AND DESCRIPTION:
images are not of good quality, due to both
hardware and software inadequacies; thus,
they have to be enhanced and improved before
other analysis can be performed on them. XII. REAL-TIME IMAGE PROCESSING:
In many of the techniques considered
IX. CONVOLUTION MASKS: so far, the image is digitized and stored before
A mask may be used for many processing. In other situations, although the
different purposes, including filtering image is not stored, the processing routines
operations and noise reduction. Noise and require long computational times before they
Edges produces higher frequencies in the are finished. This means that, in general, there
spectrum of a signal. It is possible to create is a long lapse between the time and image is
masks that behave like a low pass filter, such taken and the time a result obtained. This may
that higher frequencies of an image are be acceptable in situations in which the
attenuated while the lower frequencies are not decisions do not affect the process. However,
changed very much. There by the noise is in other situations, there is a need for real-time
reduced. processing such that the results are available in
real time or in a short enough time to be
X. EDGE DETECTION: considered real time. Two different
Edge Detection is a general name for a approaches are considered for real time
class of routines and techniques that operate processing. One is to design dedicated
on an image and results in a line drawing of hardware such that the processing is fast
the image. The lines represented changes in enough to occur in real time. The other is to
values such as cross sections of planes, try to increase the efficiency of both the
intersections of planes, textures, lines, and software and the hardware and thereby reduce
colors, as well as differences in shading and processing and computational requirements.
textures. Some techniques are mathematically
oriented, some are heuristic, and some are 4. PAPER
descriptive. All generally operate on the Here we want to present some of the
differences between the gray levels of pixels applications of Image Processing in some
or groups of pixels through masks or fields where it is applied like Robotics,
thresholds. The final result is a line drawing or Medical field and common uses…
similar representation that requires much less
memory to be stored, is much simpler to be A. APPLICATION 1:
processed, and saves in computation and Image Processing is vastly being
storage costs. Edge detection is also necessary implemented in Vision Systems in Robotics.
in subsequent process, such as segmentation Robots capture the real time images using
and object recognition. Without edge cameras and process them to fulfill the desired
detection, it may be impossible to find action.
overlapping parts, to calculate features such as A simple application in robotics using
a diameter and an area or to determine parts by Vision Systems is a robot hand-eye
region growing. coordination system. Consider that the robot’s
task is to move an object from one point to
XI. IMAGE DATA COMPRESSION: another point. Here the robots are fixed with
Electronic images contain large cameras to view the object which is to be
amounts of information and thus require data moved. The hand of the robot and the object
transmission lines with large bandwidth that is to be captured are observed by the
capacity. The requirements for the temporal cameras, which are fixed to the robot in
and spatial resolution of an image, the number position, this real time image is processed by
of images per second, and the number of gray the image processing techniques to get the
levels are determined by the required quality actual distance between the hand and the
of the images. Recent data transmission and object. Here the base wheel of the robot’s
storage techniques have significantly hand is rotated through an angle which is
improved image transmission capabilities, proportional to the actual distance between
including transmission over the Internet. hand and the object. Here a point in the target
is obtained by using the Edge Detection
Technique. The operation to be performed is
controlled by the micro-controller, which is
6. DISADVANTAGES
connected to the ports of the fingers of the
robot’s hand. Using the software programs the
operations to be performed are assigned keys • A Person needs knowledge in many
from the keyboard. By pressing the relative fields to develop an application / or
key on the keyboard the hand moves part of an application using image
appropriately. processing.
Here the usage of sensors/cameras and • Calculations and computations are
Edge Detection technique are related to Image difficult and complicated so needs an
Processing and Vision Systems. By this expert in the field related. Hence it’s
technique the complexity of using manual unsuitable and unbeneficial to ordinary
sensors is minimized to a great extent and programmers with mediocre
thereby sophistication is increased. Hence knowledge.
image processing is used here in the study of
robotics. 7. CONCLUSION
It’s a critical study, which plays a vital
B. APPLICATION 2: role in modern world as it is involved with
In the field of Medicine this is highly advanced use of science and technology. The
applicable in areas like Medical imaging, advances in technology have created
Scanning, Ultrasound and X-rays etc.,. Image tremendous opportunities for Vision System
Processing is rapidly used for MRI SCAN and Image Processing. There is no doubt that
(Magnetic Resonance Imaging) and CT SCAN the trend will continue into the future. from
(Computer Tomography). Tomography is an the above discussion we can conclude that this
imaging technique that generates an image of field has relatively more advantages than
a thin cross sectional slice of a test piece. disadvantages and hence is very useful in
varied branches.
5. ADVANTAGES
• In medicine by using the Image
Processing techniques the 8. REFERENCES
sophistication has increased. This lead
to technological advancement.  INTRODUCTION TO ROBOTICS,
• Vision Systems are flexible, ANALYSIS, SYSTEMS,
inexpensive, powerful tools that can be APPLICATIONS - SAEED B. NIKU
used with ease.
• In Space Exploration the robots play  INTRODUCTION TO DIGITAL
vital role which in turn use the image IMAGE PROCESSING – ANIL
processing techniques. K.JAIN
• Image Processing is used for
Astronomical Observations.  DIGITAL MAGE PROCESSING -
• Also used in Remote Sensing, RAFAEL C. GONZALEZ AND
Geological Surveys for detecting RICHARD E. WOODS, ADDISON
mineral resources etc. WESLEY 1993.
• Also used for character recognizing
techniques, inspection for
abnormalities in industries.