International Journal of Trend in Scientific

Research and Development (IJTSRD)

International Open Access Journal
ISSN No: 2456 - 6470 | www.ijtsrd.com | Volume - 2 | Issue – 5

Detection and Rectification ooff Distorted Fingerprints

M. Ramesh Kumari
Lecturer (Senior Grade), Department of Computer Science,
Ayyanadar Janaki Ammal Polyechnic College, Sivakasi
Sivakasi, Tamil Nadu,, India

ABSTRACT
One of the open come outs in fingerprint confirmation broader context, it implies digital processing of any
is the lack of robustness against image quality two dimensional data. A digital image is an array of
degradation. Poor-quality
quality images result in specious real or complex numbers represented by a finite
and missing features, thus degrading the performance number of bits. An image given in the form of a
of the overall system. Therefore, it is very important transparency, slide, photograph or an X-ray
X is first
for a fingerprint acknowledgement system to estimate digitized and stored as a matrix of binary digits in
the quality and validity of the captured fingerprint computer memory. This digitized image can then be
images. Also the elastic distortion of fingerprints is processed and/or displayed on a high-resolution
high
one of the major causes for false non--match. While television monitor. For display, the image is stored in
this problem impacts all fingerprint acknowledged a rapid-access
access buffer memory, which refreshes the
applications, it is especially unsafe in negative monitor at a rate of 25 frames per second to produce a
recognition applications, such as watch list and visually continuous display.
reduplication applications. In such applications,
malicious users may purposely distort their 1.1.1 THE IMAGE PROCESSING SYSTEM
fingerprints to elude identification. In the existing
approaches by matching the different dataset by
plotting Rigid Core Delta point to be specified as the
finger print recognition. In this paper, we proposed
novel algorithms to detect and rectify skin distortion
based on a individual fingerprint image. In this
process the detection is based on the patches based
approaches. In these approaches the patches is defined
as the rectangular position the distortion is not
detected at the specified fingerprint. The patches size
is varied the distortion is detection and a SVM
classifier is prepared to perform the classification Figure 1: Block Diagram for Image Processing
task. Distortion rectification (distortion field System
estimation) is considered as a regression problem,
where the inputt used is a distorted fingerprint and the DIGITIZER
output is the distortion less field. A digitizer converts an image into a numerical
representation suitable for input into a digital
Keywords: fingerprint, delta point, SVM, Matlab computer. Some common digitizers are
1. Microdensitometer
1. INTRODUCTION 2. Flying spot scanner
1.1 GENERAL 3. Image dissector
The term digital image refers to processing of a two 4. Videocon camera
dimensional picture by a digital computer. In a 5. Photosensitive solid- state arrays.

@ IJTSRD | Available Online @ www

www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug
Aug 2018 Page: 1835
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
IMAGE PROCESSOR
MASS STORAGE
An image processor does the functions of image The secondary storage devices normally used are
acquisition, storage, preprocessing, segmentation, floppy disks, CD ROMs etc.
representation, recognition and interpretation and
HARD COPY DEVICE
finally displays or records the resulting image. The The hard copy device is used to produce a
following block diagram gives the fundamental permanent
anent copy of the image and for the storage
sequence involved in an image processing system. of the software involved.

OPERATOR CONSOLE
The operator console consists of equipment and
arrangements for verification of intermediate
results and for alterations in the software as and
when require. The operator is also capable of
checking for any resulting errors and for the entry
of requisite data.

Figure 2: Block Diagram of Fundamental Sequence 1.1.2 IMAGE PROCESSING FUNDAMENTAL

Involved In an Image Processing System Digital image processing refers processing of the
image in digital form. Modern cameras may directly
As detailed in the diagram, the first step in the take the image in digital
igital form but generally images
process is image acquisition by an imaging sensor in are originated in optical form. They are captured by
conjunction with a digitizer to digitize the image. The video cameras and digitalized. The digitalization
next step is the preprocessing
ocessing step where the image is process includes sampling, quantization. Then these
improved being fed as an input to the other processes. images are processed by the five fundamental
Preprocessing typically deals with enhancing, processes, at least any
y one of them, not necessarily all
removing noise, isolating regions, etc. Segmentation of them.
partitions an image into its constituent parts or
objects. The output of segmentation is usually raw IMAGE PROCESSING TECHNIQUES
pixel data, which consists of either the boundary of This section gives various image processing
the region or the pixels in the region themselves. techniques
Representation is the process of transforming the raw
pixel data into a form useful for subsequent
processing by the computer.
puter. Description deals with
extracting features that are basic in differentiating one
class of objects from another. Recognition assigns a
label to an object based on the information provided
by its descriptors. Interpretation involves assigning
meaning to an ensemble of recognized objects. The
knowledge about a problem domain is incorporated
into the knowledge base. The knowledge base guides
the operation of each processing module and also
controls the interaction between the modules. Not all
modules needd be necessarily present for a specific
function. The composition of the image processing
system depends on its application. The frame rate of
the image processor is normally around 25 frames per Figure 3 : Image Processing Techniques
second.

DIGITAL COMPUTER IMAGE ENHANCEMENT
Mathematical processing of the digitized image Image enhancement operations improve the qualities
such as convolution, averaging, addition, of an image like improving the image’s contrast and
subtraction, etc. are done by the computer. brightness characteristics, reducing its noise content,

@ IJTSRD | Available Online @ www

www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug
Aug 2018 Page: 1836
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
or sharpen the details. This just enhances the image MEDICAL APPLICATIONS
and reveals the same information in more In medical applications, one is concerned with
understandable image. It does not add any information processing of chest X-rays, cineangiograms,
to it. projection images of transaxial tomography and other
medical images that occur in radiology, nuclear
IMAGE RESTORATION magnetic resonance (NMR) and ultrasonic scanning.
Image restoration like enhancement improves the These images may be used for patient screening and
qualities of image but all the operations are mainly monitoring or for detection of tumors’ or other disease
based on known, measured, or degradations of the in patients.
original image. Image restorations are used to restore
images with problems such as geometric distortion, SATELLITE IMAGING
improper focus, repetitive noise, and camera motion. Images acquired by satellites are useful in tracking of
It is used to correct images for known degradations. earth resources; geographical mapping; prediction of
agricultural crops, urban growth and weather; flood
IMAGE ANALYSIS and fire control; and many other environmental
Image analysis operations produce numerical or applications. Space image applications include
graphical information based on characteristics of the recognition and analysis of objects contained in image
original image. They break into objects and then obtained from deep space-probe missions.
classify them. They depend on the image statistics.
Common operations are extraction and description of COMMUNICATION
scene and image features, automated measurements, Image transmission and storage applications occur in
and object classification. Image analyze are mainly broadcast television, teleconferencing, and
used in machine vision applications. transmission of facsimile images for office
automation, communication of computer networks,
IMAGE COMPRESSION closed-circuit television based security monitoring
Image compression and decompression reduce the systems and in military communications.
data content necessary to describe the image. Most of
the images contain lot of redundant information, RADAR IMAGING SYSTEMS
compression removes all the redundancies. Because Radar and sonar images are used for detection and
of the compression the size is reduced, so efficiently recognition of various types of targets or in guidance
stored or transported. The compressed image is and maneuvering of aircraft or missile systems.
decompressed when displayed. Lossless compression
preserves the exact data in the original image, but DOCUMENT PROCESSING
Lossy compression does not represent the original It is used in scanning, and transmission for converting
image but provide excellent compression. paper documents to a digital image form, compressing
the image, and storing it on magnetic tape. It is also
IMAGE SYNTHESIS used in document reading for automatically detecting
Image synthesis operations create images from other and recognizing printed characteristics.
images or non-image data. Image synthesis operations
generally create images that are either physically DEFENSE/INTELLIGENCE
impossible or impractical to acquire. It is used in reconnaissance photo-interpretation for
automatic interpretation of earth satellite imagery to
APPLICATIONS OF DIGITAL IMAGE look for sensitive targets or military threats and target
PROCESSING acquisition and guidance for recognizing and tracking
Digital image processing has a broad spectrum of targets in real-time smart-bomb and missile-guidance
applications, such as remote sensing via satellites and systems.
other spacecrafts, image transmission and storage for
business applications, medical processing, radar, PROJECT DESCRIPTIONS:
sonar and acoustic image processing, robotics and The Goal of the Project a distorted fingerprint can be
automated inspection of industrial parts. thought of being generated by applying an unknown

@ IJTSRD | Available Online @ www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug 2018 Page: 1837
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
distortion field d to the normal fingerprint, which is fingerprints. It is based on the patches based
also unknown. If we can estimate the distortion field d classifications. When the patches length is varied on
from the given distorted fingerprint, we can easily the threshold value is to be defined as a distorted
rectify it into the normal fingerprint by applying the fingerprint. When the patches length is based on the
inverse of d. So we need to address a regression threshold value is to be defined as the distortion less
problem, which is quite difficult because of the high fingerprint. This is based on the characteristics
dimensionality of the distortion field (even if we use a extracted from the orientation field and minutiae
block-wise distortion field). In this paper, a nearest perform. SVM classifier is prepared to perform the
neighbor regression approach is used for this task. classification task. Distortion rectification (distortion
The proposed distorted fingerprint rectification field estimation) is considered as a regression
algorithm consists of an offline stage and an online problem, where the input used is a distorted
stage. In the offline stage, a database of distorted fingerprint and the output is the distortion less field.
reference fingerprints is generated by transforming The proposed algorithm is based on the characteristics
several normal reference fingerprints with various extracted from the orientation field and minutiae
distortion fields sampled from the statistical model of perform or satisfy the three required requirements for
distortion fields. In the online stage, given a distorted alteration detection algorithm: 1) speedy operational
input fingerprint, we retrieve its nearest neighbor in time, 2) Huge true positive rate at small false positive
the distorted reference fingerprint database and then rate, and 3) Ease of integration into AFIS.
use the inverse of the corresponding distortion field to
rectify the distorted input fingerprint. Fingerprint rectification algorithm consists of an
offline stage and an online stage. In the offline stage,
2. SYSTEM ANALYSIS a database of distorted reference fingerprints is
In Existing System, Distortion detection is believed as generated by transforming several normal reference
a two class classification problem, where the fingerprints with various distortion fields sampled
registered ridge orientation map and period map of a from the statistical model of distortion fields.
particular fingerprint are utilized as the feature vector
and a classifier is trained to execute the classification The proposed distortion rectification algorithm by
task. Distortion correction (or equivalently distortion performs well by performing matching experiments
field estimation) is considered as a regression on various databases.
problem, the input is a distorted fingerprint and the
output is the distortion field. To solve this problem, a The proposed algorithm can improve recognition rate
database of several distorted reference fingerprints of distorted fingerprints evidently.
and accompanying distortion fields is made in the
offline stage, and then in the online stage, the closest 3. SYSTEM IMPLEMENTATION
neighbor of the input fingerprint is found in the MODULES
database of distorted reference fingerprints and the
Detect distorted fingerprints
corresponding distortion field is used to correct the
Extracted Distorted Fingerprint
input fingerprint. They require special force sensors
fingerprint Verification algorithm
or fingerprint sensors with video capturing capability.
distorted fingerprint rectification algorithm
They cannot detect distorted fingerprint images in
existing fingerprint databases. They cannot detect 3.1 Detect distorted fingerprints
fingerprints distorted before pressing on the sensor. Fingerprint distortion detection can be considered as a
However, allowing larger distortion in matching will two class classification problem. We have used the
inevitably result in higher false match rate. For registered ridge orientation map and period map as the
example, if we increased the bounding zone around a feature vector, which is further classified by SVM
minutia, many non-mated minutiae will have a chance classifier. An input fingerprint image which is
to get paired. In addition, allowing larger distortion in provided is normalized by cropping or cutting a
matching will also slow down the matching speed. rectangular region of the input image fingerprint,
which is located at the center of the fingerprint and
In Proposed System was evaluated based on finger aligned along with the longitudinal direction of the
level. At the finger level, we estimate the performance fingerprints, using the NIST Biometric Image
of differentiating between natural and changed Software (NBIS). This step insures that the features

@ IJTSRD | Available Online @ www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug 2018 Page: 1838
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
extracted in the subsequent steps are invariant with field). We use nearest neighbor regression approach
respect to translation and rotation of finger. The for this project.
orientation field of the fingerprint
ngerprint is estimated using
the gradient-based
based method. The starting orientation 4. SYSTEM DESIGN
field is smoothed moderating filter, followed by 4.1 SYSYTEM ARCHITECTURE DIAGRAM
moderating the orientations in pixel blocks. A
foreground mask is earn by measuring the dynamic
range of gray values of the fingerprint
ngerprint image in local
blocks and morphological process for filling holes and
removing isolated blocks is performed. The
orientation field is near by a polynomial model to
obtain. The error map is counted as the absolute
difference in-between and used to construct the
feature vector.

3.2 Extracted Distorted Fingerprint:

To extract Distorted based on the Patches Conversion.
The fingerprint images is Converted into the block
based representation. Each and every block is Varied
depends upon the threshold values
alues is based on the Figure 4: System Architecture Design
distorted fingerprint. The block is not varied upon the
threshold values it is defined as the distortion less 5. SYSTEM STUDY
fingerprint. And the based on the each block to 5.1 FEASIBILITY STUDY
extraction distortion part. The feasibility of the project is analyzed in this phase
and business proposal is put forth with a very general
3.3 Fingerprint Verification algorithm plan for the project and some cost estimates. During
The most widely used minutiae-based based fingerprint system analysis the feasibility study of the proposed
matching method, In this methodology, a minutia in system is to be carried out. This is to ensure that the
the fingerprint implies the ridge characteristics such proposed system is not a burden to the company. For
as ridge ending or ridge bifurcation. Almost all the feasibility analysis, some understanding of the major
fingerprint recognition systems usage minutiae for requirements for the system is essential.
matching.
hing. The abnormality observed in orientation
field also noted that minutiae distribution of altered Three key considerations involved in the feasibility
fingerprints often differs from that of natural analysis are
fingerprints. On the basis of minutiae extracted from a
ECONOMICAL FEASIBILITY
fingerprint by the open source minutiae extractor iin
TECHNICAL FEASIBILITY
NBIS, a minutiae density map is composed by using
SOCIAL FEASIBILITY
the Parzen window method containing uniform kernel
function. 5.1.1 ECONOMICAL FEASIBILITY
This study is carried out to check the economic
3.4 Distorted
orted fingerprint rectification impact that the system will have on the organization.
algorithm Thee amount of fund that the company can pour into
A distorted fingerprint can be thought of being the research and development of the system is limited.
generated by applying a strange distortion field to the The expenditures must be justified. Thus the
normal fingerprint, which is also strange. If we can developed system as well within the budget and this
calculate the distortion field from the given distorted was achieved because most of the technologies used
fingerprint, we can easily correct it into the normal are freely
reely available. Only the customized products had
fingerprint by applying the inverse of d. So we need to be purchased.
to turn to a regression problem,
blem, which is quite hard
because of the high dimensionality of the distortion 5.1.2 TECHNICAL FEASIBILITY
field (although if we use a block-wise wise distortion This study is carried out to check the technical
feasibility, that is, the technical requirements of the

@ IJTSRD | Available Online @ www

www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug
Aug 2018 Page: 1839
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
system. Any system developed must not have a high 6.4 Rectify Distorted Fingerprint Based on the
demand on the available technical resources. This will Patches
lead to high demands on the available technical
resources. This will lead to high demands being
placed on the client. The developed system must have
a modest requirement, as only minimal or null
changes are required for implementing this system.

5.1.3 SOCIAL FEASIBILITY

The aspect of study is to check the level of acceptance
of the system by the user. This includes the process of
training the user to use the system efficiently. The
user must not feel threatened by the system, instead 6.5 Oriented Field Representation
must accept it as a necessity. The level of acceptance
by the users solely depends on the methods that are
employed to educate the user about the system and to
make him familiar with it. His level of confidence
must be raised so that he is also able to make some
constructive criticism, which is welcomed, as he is the
final user of the system.

6. SCREENSHOT
6.1 Original Images:
6.6 Minutiae Based Quality Measurement

6.2 Find Distorted Fingerprint:

7. CONCLUSION
False non-match frequency of fingerprint matchers is
relatively high in severely distorted fingerprints. It
creates a security hole in automatic fingerprint
detection systems that could be used by criminals and
terrorists. So, building up of fingerprint distortion
6.3 Extracted Distorted Fingerprint scrutiny and reformation algorithms to fill the hole is
a must. The paper illustrates a new distorted
fingerprint detection and rectification algorithm.
Distortion detection is done by the use of registered
ridge orientation map and period map of a fingerprint
as the feature vector, a SVM classifier is made to
classify the input fingerprint as distorted or normal. In
distortion rectification (or distortion field estimation),
a nearest neighbor regression method is employed to
anticipate the distortion field from the input distorted

@ IJTSRD | Available Online @ www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug 2018 Page: 1840
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
fingerprint , later the inverse of the distortion field is 9. F. Alonso-Fernandez, J. Fierrez-Aguilar, J.
used to change the distorted fingerprint into a normal Ortega-Garcia, J. Gonzalez-Rodriguez, H.
one. The experimental results on FVC2004 DB1, Fronthaler, K. Kollreider, and J. Bigun, “A
Tsinghua DF database, and NIST SD27 database €comparative study of fingerprint image-quality
show that the proposed algorithm can enhance rate of estimation methods,” IEEE Trans. Inf. Forensics
identification of distorted fingerprints unmistakably. Security, vol. 2, no. 4, pp. 734–743, Dec. 2007.
10. J. Fierrez-Aguilar, Y. Chen, J. Ortega-Garcia, and
8. REFERENCES A. K. Jain, “Incorporating image quality in multi-
1. X. Si, J. Feng, and J. Zhou, “Detecting fingerprint algorithm fingerprint verification, ”in Proc. Int.
distortion from asingle image,” in Proc. IEEE Int. Conf. Biometrics, 2006, pp. 213–220.
Workshop Inf. Forensics Security, 2012, pp. 1–6.
11. L. Hong, Y. Wan, and A. K. Jain, “Fingerprint
2. D. Maltoni, D. Maio, A. K. Jain, and S. Prabhakar,
image enhancement: Algorithm and performance
Handbook of Fingerprint Recognition, 2nd ed. evaluation,” IEEE Trans. Pattern Anal. Mach.
Berlin, Germany: Springer-Verlag, 2009.
Intell., vol. 20, no. 8, pp. 777–789, Aug. 1998.
3. FVC2006: The fourth international fingerprint
12. S. Chikkerur, A. N. Cartwright, and V.
verification competition. (2006). [Online].
Govindaraju, “Fingerprint enhancement using
Available: http://bias.csr.unibo.it/fvc2006/
STFT analysis,” Pattern Recognit., vol. 40, no. 1,
4. V. N. Dvornychenko, and M. D. Garris, pp. 198–211, 2007.
“Summary of NIST latentfingerprint testing
13. F. Turroni, R. Cappelli, and D. Maltoni,
workshop,” Nat. Inst. Standards Technol., “Fingerprint enhancementusing contextual
Gaithersburg, MD, USA, Tech. Rep. NISTIR iterative filtering,” in Proc. Int. Conf. Biometrics,
7377, Nov. 2006. 2012, pp. 152–157.
5. Neurotechnology Inc., VeriFinger. (2009).
14. J. Feng, J. Zhou, and A. K. Jain, “Orientation field
6. L. M. Wein and M. Baveja, “Using fingerprint estimation forlatent fingerprint enhancement,”
image quality toimprove the identification IEEE Trans. Pattern Anal. Mach. Intell., vol. 35,
performance of the U.S. visitor andimmigrant no. 4, pp. 925–940, Apr. 2013.
status indicator technology program,” Proc. Nat. 15. X. Yang, J. Feng, and J. Zhou, “Localized
Acad. Sci. USA, vol. 102, no. 21, pp. 7772–7775, dictionaries based orientation field estimation for
2005. latent fingerprints,” IEEE Trans. Pattern Anal.
7. S. Yoon, J. Feng, and A. K. Jain, “Altered Mach. Intell., vol. 36, no. 5, pp. 955–969, May
fingerprints: Analysisand detection,” IEEE Trans. 2014.
Pattern Anal. Mach. Intell., vol. 34,no. 3, pp. 451–
464, Mar. 2012.
8. E. Tabassi, C. Wilson, and C. Watson,
“Fingerprint image quality, ”Nat. Inst. Standards
Technol., Gaithersburg, MD, USA, Tech. Rep.
NISTIR 7151, Aug. 2004.

@ IJTSRD | Available Online @ www. ijtsrd. com | Volume – 2 | Issue – 5 | Jul-Aug 2018 Page: 1841