Professional Documents
Culture Documents
2018 - DetectingFakeIris&VijayKumar - AnujKumar - ManishMahajan
2018 - DetectingFakeIris&VijayKumar - AnujKumar - ManishMahajan
2018 - DetectingFakeIris&VijayKumar - AnujKumar - ManishMahajan
Digital Investigation
journal homepage: www.elsevier.com/locate/diin
a r t i c l e i n f o a b s t r a c t
Article history: Iris recognition is an automated method of biometric identification that uses mathematical pattern-
Received 7 December 2017 recognition techniques on video images of the irises of an individual's eyes, whose complex random
Received in revised form patterns are unique and can be seen from some distance. Now days, Iris is being used widely by several
15 February 2018
organizations, including governments, for identification and authentication purposes. Aadhar, India's UID
Accepted 12 March 2018
Available online 20 March 2018
project uses Iris scan along with fingerprints to uniquely identify people and allocate a Unique Identi-
fication Number. Most of the work done in the area of Iris pattern recognition systems emphasizes only
on matching of the patterns with the stored templates. Security aspects of the system are still unex-
Keywords:
Iris
plored. The available security algorithms provide only some cryptographic solutions that keeps the
Template template database in a secret cryptographic form. We successfully enhanced the detection of fake iris
Luster images and add the provision of detection of false of scanned iris images as template. This enhanced
Cryptography significantly the performance of the system in terms of security and reliability. We use Flash and motion
Biometrics detection of natural eye to detect the liveliness of real iris images before matching from stored templates.
UID © 2018 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.diin.2018.03.002
1742-2876/© 2018 Elsevier Ltd. All rights reserved.
98 V.K. Sinha et al. / Digital Investigation 25 (2018) 97e104
The principle
prepared ahead of time and will inevitably build the season of iris propelled elements of cryptography, the system gives higher un-
acknowledgment. wavering quality and security in Iris acknowledgment (Fig. 5).
In this research paper, we propose an intelligent fake iris Gaganpreet et al. (Kaur et al., 2010) have proposed an improved
detection system in light of wavelet parcel change together with iris acknowledgment technique which overcomes issues like
Light Support Vector Machine which can identify the paper printed reducing so as to expand the rate the multifaceted nature of the
iris successfully. Wavelet bundle change is firstly d to remove the system. For the most part, three stages are taken after while
elements. At that point by means of Light Support Vector Machine working with the iris acknowledgment framework viz. preprepar-
is utilized to characterize fake irises from genuine ones. We showed ing, highlight extraction and acknowledgment stage. This paper
that fake iris images have lacking in reflection images and dull introduces a robotized and novel iris acknowledgment framework
reflection of Flash lights. where general computational match velocity is lessened while as
yet having a precision of 99.38% and low FAR.
Literature survey Ifeanyi Ugbaga N. Kole et al. (Nkole and Sulong, 2011) have
proposed another technique for division of the iris picture by uti-
It has been long realized that our iris examples (like retina ex- lizing 8 neighborhood administrators. This picture is then bolstered
amples and fingerprints) are one of a kind and these can be utilized into a circle Hough Transform to upgrade iris division which is the
for individual recognizable proof. Notwithstanding, this could be most challenged issue in the iris acknowledgment framework. For
misused just in late 1980s when fitting innovation was accessible this examination 320 iris pictures from CASIA standard dataset
for executing such a framework. The beginning of a dependable iris were utilized. The dispatch demonstrates a higher exactness rate.
acknowledgment framework can be followed to 1987 when two P. Lorrentz et al. (Lorrentzet al., 2009) have explored how hu-
ophthalmology educators, Leonard Flom (MD, New York Univer- man distinguishing proof and personality confirmation can be
sity) and AranSafir (MD, University of Connecticut) were issued a performed by the use of FPAG based weightless neural system to
first of its kind wide patent # 4,641,349 entitled “Iris Recognition the iris biometric methodology. The human iris is handled for
Technology”. Along these lines, Dr. John Daugman (Harvard Com- highlight vectors which are utilized for arrangement of availability
puter Science workforce) was asked by the two eye specialists to amid learning and resulting acknowledgment.
compose a calculation for their idea based upon a broad arrange- Nithyanaudam S. et al. (Nithyanandam et al., 2011) utilize a
ment of high determination iris photographs supplied to him by Dr. Canny Edge Detection Scheme and a Circular Hough Transform to
Flom from his volunteer private patients. Quite a long while later, recognize iris limits in the eye's advanced picture.
Daugman got a technique patent for the calculation. The three In any case, to the best of our insight, the utilization of Flash
people then established “Iridian Technologies Inc.” for further location (of normal eyes) and utilization of double iris rather than a
adding to the idea. This strategy in view of Daugman calculation solitary iris (i.e. whether the specific iris picture is from the left or
was authorized to a few partnerships for commercializing the right side eye of the individual) have not been utilized till now for
framework (Donnelly, 2001e11). location/confirmation of the individual under thought.
Upon the termination of the Flom/Safir patent in 2008, different
calculations were protected; some of them were better than Proposed work
Daugman's. Some of these are as a rule further created/marketed by
US Govt. Organizations (Flom and Safir, 1987). For template making, the high resolution scanned images of the
On account of Daugman calculation, a Gabor wavelet change iris are processed by various computer subroutines.
(Hao et al., 2006) is utilized on the pixels of the human iris. This First, an iris recognition algorithm (subroutine) is used to
outcomes in a progression of complex numbers which convey the localize the inner and outer boundaries of the iris. Parts of eyelids,
adequacy and stage data about the iris design. Most abundances eye lashes and specular reflections which often occlude parts of the
data is disposed of. This guarantees the layout remains to a great iris are detected and excluded from the images scanned.
extent unaffected by changes in enlightenment or camera pick up The set of pixels containing only the iris are then normalized to
and adds to the long haul ease of use of the biometric format. This compensate for pupil dilation or constriction. The information
format can then use to look at the sweeps of the iris from diverse stored in each pixel is then encoded to obtain a set of complex
persons for recognizable proof. numbers that carry the amplitude and phase information about iris
From that point forward, a ton of innovative work has been done patterns. The amplitude information thus stored may be discarded
in this field. The anxiety has been on security and distinguishing to ensure that the template remains largely unaffected by changes
proof of the iris picture when contrasted and the put away layout. in illumination or camera gain (contrast). This contributes to the
Utilization of Error Correcting Codes (ECC) has been utilized to sift long term usability of the biometric template.
through clamors and mistakes which may happen in the caught Various transformations and encoding techniques are used in
pictures because of burst or foundation blunders which can this process of template making to ensure security and longevity of
creepein in each new iris filter. Sim Hieu Moi et al. (2009) have the template (Istanbul, 2009).
utilized Error Correcting Codes to dispense with these commotions
and burst blunders in the distinguishing proof of the iris filters. Process of identification/verification
They have utilized two distinctive separation metric capacities in
the format coordinating procedure. For identification (one to many template matching) or verifi-
The Chinese Academy of Sciences (CASIA) have built up an Iris cation (one to one template matching), a template created by im-
Database (form 1.0) of 756 iris tests taken from 108 persons (with aging an iris is compared with stored template(s) in a database.
seven unique pictures for every individual) These pictures are 8-bit In this process, the steps in template making are repeated and
dark level JPEG documents (Tan et al., 2010). Iris formats are made this digital image (template of the person to be identified/verified)
by iris division, standardization and highlight extraction procedure is then compared with the stored template in the database. Various
to deliver the iris twofold code which is then encoded utilizing criterions may be employed for comparison purposes. In one
Reed Solomon Code. This RS Code is then encoded utilizing a secret commonly used method, if the Hamming distance is below the
word to frame a Cipher content which is then put away in the decision threshold, a positive identification is made (Ray, 2006).
database as a format (Fig. 4). With the utilization of division and Like everything else, this system has advantages and drawbacks.
100 V.K. Sinha et al. / Digital Investigation 25 (2018) 97e104
Advantages of iris identification/verification The iris has a fine texture thatd like fingerprintsdis deter-
mined randomly during embryonic gestation. Like the fingerprints,
The iris of the eye has been described as the ideal part of the it is very hard (if not impossible) to prove that the iris is unique.
human body for biometric identification for several reasons: However, there are so many factors that go into the formation of
It is an internal organ that is well protected against damage and these textures (the iris and fingerprint) that the chance of false
wear by a highly transparent and sensitive membrane (the cornea). matches for either is extremely low. Even genetically identical in-
This distinguishes it from fingerprints, which can be difficult to dividuals have completely independent iris textures.
recognize after years of certain types of manual labor. An iris scan is similar to taking a photograph and can be per-
The iris is mostly flat, and its geometric configuration is only formed from about 10 cm to a few meters away. There is no need for
controlled by two complementary muscles (the sphincter papillae the person being identified to touch any equipment that has
and dilator papillae) that control the diameter of the pupil. This recently been touched by a stranger, thereby eliminating an ob-
makes the iris shape far more predictable than, for instance, that jection that has been raised in some cultures against fingerprint
of the face. scanners, where a finger has to touch a surface, or retinal scanning,
V.K. Sinha et al. / Digital Investigation 25 (2018) 97e104 101
where the eye must be brought very close to an eyepiece (like (voluntary eye movement) tracking eye while text is read, etc
looking into a microscope). (Mosca et al., 2003).
The commercially deployed iris recognition algorithm, John Testing for retinal retro reflection (red-eye effect). Testing for
Daugman's Iris Code, has an unprecedented false match rate (better reflections from the eye's four optical surfaces (front and back of
than 10 11 if a Hamming distance threshold of 0.26 is used, both cornea and lens) to verify their presence, position and shape.
meaning that up to 26% of the bits in two Iris Codes are allowed to Using 3D imaging (e.g., stereo cameras) to verify the position and
disagree due to imaging noise, reflections, etc., while still declaring shape of the iris relative to other eye features.
them to be a match). (Hao et al., 2006).
While there are some medical and surgical procedures that can Proposed research work
affect the color and overall shape of the iris, the fine texture re-
mains remarkably stable over many decades. Some iris identifica- The present research is made to improve the reliability and se-
tions have succeeded over a period of about 30 years. curity of the iris detection.
System by detect and distinguish fake scanned iris images with
Problem statement real human iris images. As Iris scanners can be made fool by pre-
senting a scanned high resolution scanned iris image to make scams.
Many commercial Iris scanners can be easily fooled by a high
quality image of an iris or face in place of the real thing. The Objectives of research
scanners are often tough to adjust and can become bothersome for
multiple people of different heights to use in succession. The ac- The main aim & objectives of current research work are to
curacy of scanners can be affected by changes in lighting. Iris enhance the security and intelligence of existing Iris recognition
scanners are significantly more expensive than some other forms of system which includes following objectives:
biometrics, password or proxy card security systems. To study the existing available iris recognition system.
Iris scanning is a relatively new technology and is incompatible To improve the security level of existing iris recognition system
with the very substantial investment that the law enforcement and by taking both iris together.
immigration authorities of some countries have already made into To test and evaluate the proposed system for efficiency with
fingerprint recognition. existing techniques.
Iris recognition is very difficult to perform at a distance larger
than a few meters and if the person to be identified is not coop- Research methodology & tools
erating by holding the head still and looking into the camera.
However, several academic institutions and biometric vendors are We used MATLAB for result analysis and algorithm imple-
developing products that claim to be able to identify subjects at mentations. Statistical tools are used for data analysis.
distances of up to 10 m (“standoff iris” or “iris at a distance” as well
as “iris on the move” for persons walking at speeds up to 1 m/sec). Iris enrollment process (template making)
(Fancourt et al., 2005).
As with other photographic biometric technologies, iris recog- Following steps will be used in template making (Fig. 4). Take
nition is susceptible to poor image quality, with associated failure images of both the irises of the volunteers at five different angles
to enroll rates. viz. in the frontal normal position.
As with other identification infrastructure (national residents At 45 angle (Left and Right) in the horizontal plane.
databases, ID cards, etc.), civil rights activists have voiced concerns At 45 angle (Up and Down) in the vertical plane.
that iris-recognition technology might help governments to track The enrollment steps are described below (Moi et al., 2010):
individuals beyond their will. As with most other biometric iden-
tification technology, a still not satisfactorily solved problem with Step 1: Irises extracted through iris Segmentation by using
iris recognition is the problem of live-tissue verification. The reli- thresholding (Tobias and Seara, 2002), Iris Normalization by
ability of any biometric identification depends on ensuring that the using normalization algorithms (Schadt et al., 29 JAN 2002) and
signal acquired and compared has actually been recorded from a Feature Extraction process by RGB(Color) to Grey conversion,
live body part of the person to be identified and is not a manu- Thinning, Smoothing, etc. to generate the iris template and to
factured template. Many commercially available iris-recognition produce the iris binary code (Johnson, 2006).
systems are easily fooled by presenting a high-quality photograph Step 2: The binary code for the iris image will then undergo the
of a face instead of a real face, which makes such devices unsuitable Reed Solomon Code Encoding Process i.e. Grey to binary con-
for unsupervised applications, such as door access-control systems. version (Forney, 1965).
The problem of live-tissue verification is less of a concern in su- Step 3: The RS Code is then encrypted with the enrollment pass-
pervised applications (e.g., immigration control), where a human word using Advanced Encryption Standard Cryptography Algo-
operator supervises the process of taking the picture. rithm to generate a cipher text (Courtois and Pieprzyk, 2002).
Methods that have been suggested to provide some defense Step 4: The generated cipher texts then stored in the database as
against the use of fake eyes and irises. These include: Changing a template. The final extracted Iris pattern templates are
ambient lighting during the identification (switching on a bright encrypted by using AES encryption algorithm.
lamp), such that the pupillary reflex can be verified and the iris
image be recorded at several different pupil diameters. Analyzing Proposed algorithm
the 2D spatial frequency spectrum of the iris image for the peaks
caused by the printer dither patterns found on commercially Step 1: Start iris diagnosis
available fake-iris contact lenses. Analyzing the temporal frequency Step 2: Capture image
spectrum of the image for the peaks caused by computer displays. Step 3: Test Flash Level
Using spectral analysis instead of merely monochromatic cam- Step 4: If Flash light reflection is below a certain level (Dull)
eras to distinguish iris tissue from other material. Observing the detect fake Iris and stop else proceed.
characteristic natural movement of an eyeball, measuring Step 5: Test natural movement of Eye ball
102 V.K. Sinha et al. / Digital Investigation 25 (2018) 97e104
Step 6: If movement of eye ball is not natural then detect fake Firstly we conduct tests for the natural iris verification. For this
iris image and stop else proceed we tested for Flash detection using Shine Artifact algorithm (Gilat
Step 7: Conduct Template matching from stored database and second ed., 2004).
Step 8: Match left Iris with left iris template and right iris with If iris lecture detected natural then we go for next step i.e.
right iris template movement detection for a natural iris by using difference algorithm
Step 9: If Iris is matching then declare user identified and and threshold filters.
authorized else declare unidentified and unauthorized. After movement verification process we proceed for the next iris
Step 10: Exit iris diagnosis verification process (Refer to Fig. 6).
For this we repeat all the steps of iris enrollment process as
Iris verification process described above i.e.: Iris Segmentation, Iris Normalization,
Feature Extraction, Iris Template. Generation, RS Decoding,
The next stage is the verification/identification process of the Template Matching by using Hamming Distance and Weighted
target person. After iris verification process will continue. For this, I Euclidian Distance. This would be the template of the target
propose to take iris samples of volunteers (randomly from those person.
whose iris images are stored in the database as well as those whose This template of the target person is then matched with the
iris images are not in the database). Verification is proposed to be stored template in the database for verification/identification
carried out in the following steps (refer to Fig. 5). process.
Finally iris is declared matched if all the specified criteria are Daugman, J., 1999. Recognizing persons by their Iris patterns: countermeasures
against subterfuge. In: Jain, et al. (Eds.), Biometrics. Personal Identification in a
met otherwise declared not matched or false iris.
Networked Society, pp. 103e121.
Iris Verification Process 9.2 Result and Discussions. Donnelly, Ally, Sunday Nov. 18, 2001. Your eyes can tell No lies. Time Mottto 1 (1).
The above algorithm is tested and analyzed on various iris http://content.time.com/time/magazine/article/0,9171,185021,00.html.
templates. Fancourt, C., Bogoni, Luca, Hanna, Keith, Guo, Yanlin, Wildes, Richard, Takahashi, Naomi,
Jain, Uday, 2005. Iris recognition at a distance, vol. 3546. Springer.
Testing of Real vs. Fake scanned Iris detection accuracy: Feng, X., Ding, X., Wu, Y., Wang, P.S.P., 2008. Classifier combination and its application
From the above both tables of Real and Fake scanned Iris Images in iris recognition. Int. J. Pattern Recogn. Artif. Intell. 22 (3), 617e638.
we can conclude that the security algorithm introduced for Flash Flom, Safir, A., 1987. Iris Recognition System. U.S. Patent No.4641394.
Forney Jr., G., October 1965. On decoding BCH codes. IEEE Trans. Inf. Theor. 11 (4),
detection and natural movement detection are robust enough to 549e557. https://doi.org/10.1109/TIT.1965.1053825.
detect and protect the system from scammers of fake users. Gilat, Amos, 2004. MATLAB: an Introduction with Applications. In: second ed (Ed.).
John Wiley & Sons. ISBN 978-0471-69420-5.
Hao, F., Anderson, R., Daugman, J., 2006. Combining crypto with biometrics effec-
Comparison with existing method tively. IEEE Trans. Comput. 55 (no. 9), 1081e1088.
Istanbul, 2009. A frequency based encoding technique for transformation of cate-
gorical variables in mixed IVF dataset engineering in Medicine and Biology
Here, we will present a comparison between the current Society. In: EMBC 2009. Annual International Conference of the IEEE 3-6 Sept.
method and Daugman method and Xiaofu method Described for 2009 Uyar. A. Bogazici Univ., Turkey, pp. 6214e6217.
the purpose of comparison. Jain, A.K., Bolle, R.M., Pankanti, S. (Eds.), 1999. Biometrics: Personal Identification in
Networked Society. Kluwer, Norwell.
We implement his method according to the published paper.
Johnson, Stephen, 2006. Stephen Johnson on Digital Photography. O'Reilly. ISBN 0-
596-52370-X.
Conclusions & future scope Kaur, Ms Gaganpreet, Girdhar, Mr Akshay, Kaur, Ms Manvjeet, October 2010.
Enhanced Iris recognition system e an integrated approach to person identi-
fication. Article: enhanced Iris recognition system - an integrated approach to
We successfully enhanced the detection of fake iris images and person identification. Int. J. Comput. Appl. 8 (1), 1e5.
add the provision of detection of false of scanned iris images as Lee, E.C., Park, K.R., Kim, J., 2006. Fake iris detection by using purkinje image. In:
Zhang, D., Jain, A.K. (Eds.), ICB 2006. LNCS, Vol. 3832. Springer, Heidelberg,
template. This enhanced significantly the performance of the pp. 397e403, 17.Daubechies, I.: Orthonormal bases of compactly supported
system in terms of security and reliability. We use Flash and wavelets. Commun. Pure Appl. Math. XLI, 909e996(1988).
motion detection of natural eye. We successfully attained 99.98% Lorrentz, P., Howells, W.G.J., McDonald-Maier, K.D., 2009. FPGA-based
enhanced probabilistic convergent weightless Network for human Iris
accuracy at 5.2% threshold value at 10.5 cm distance from the Iris recognition. In: Proceedings of European Symposium on Artificial Neural
scanner. Networks-advances in Computational Intelligence and Iearning. Bruges,
Belgium, pp. 319e324.
Ma, L., Tan, T., Wang, Y., Zhang, D., 2003. Personal identification based on iris texture
Future scope analysis. IEEE Trans. Pattern Anal. Mach. Intell. 25 (12), 1519e1533.
Moi, Sim Hiew, Rahim, Nazeema Binti Abdul, Saad, Puteh, Sim, Pang Li,
Zakaria, Zalmiyah, Ibrahim, Subariah, 2009. Iris biometric cryptography for
This research can be further expanded for the detection of fake
identity document. In: International Conference of Soft Computing and Pattern
finger print images and other authentication systems. This research Recognition, IEEE Computer Society Conference on Dec 2009, p. 736741.
can be expanded for significant distance iris images where camera Moi, Sim Hiew, Saad, Puteh, Ibrahim, Subariah, 2010. Enhancing Iris biometric
normally not able to distinguish the level of Flash. This research can recognition system with cryptography and error correction codes :a pre-
liminary study. IEEE.
be further refined by reducing the level of threshold value by less Mosca, F., Sicignano, S., Leone, C.A., 2003. Benign positional paroxysmal vertigo:
than 5.2%. video nystagmo graphic study using rotator test. Acta Otorhinolaryngol. Ital. 23
(2), 67e72.
Nithyanandam, S., Gayathri, K.S., Priyadarshini, P.L.K., July 2011. A new IRIS
Acknowledgement normalization process for recognition system with cryptographic techniques ;
CSE department, PRIST University, Thanjavur, Tamilnadu, India. IJCSI Int. J.
Comput. Sci. Issues 8 (4), 342e344. No 1. www.IJCSI.org. ISSN (Online): 1694-
We are thankful for the Punjab technical University, Kapurthala 0814.
(India) for the opportunity provided us to conduct current this Nkole, Ifeanyi Ugbaga, Sulong, Ghazali Bin, 2013. An enhanced iris segmentation
research work on Enhancement of security and Intelligence of iris algorithm using circle hough transform. IEEE, 2011. (IJCSIT) Int. J. Comput. Sci.
Inf. Technol. 4 (1), 83e86.
Recognition system. We also grateful to the Chandigarh engineer- Park, K.R., Kim, J., 2005. A real-time focusing algorithm for iris recognition camera.
ing college, Landran, Mohali, Punjab, (India) affiliated to PTU, IEEE Trans. Syst. Man Cybern C 35 (3), 441e444.
Kapurthala for supporting our research work. The active and con- Prabhakar, S., Kittler, J., Maltoni, D., O'Gorman, L., Tan, T., 2007. Introduction to the
special issue on biometrics: progress and directions. IEEE Trans. Pattern Anal.
stant support of our research supervisors Dr. Ravinder Khanna & Dr.
Mach. Intell. 29 (4), 513e516.
Anuj Kumar Gupta & Dr. Manish Mahajan being the main source of Ray, Justin, June 25-28 2006. Efficient High Hamming Distance CRCs for Embedded
inspiration led this research a success. Networks. Carnegie Mellon University. justinr2@andrew.cmu.edu.
Schadt, Eric E., Li, Cheng, Ellis, Byron, Wong, Wing H., 29 JAN 2002. Feature
extraction and normalization algorithms for high density oligo nucleotide gene
Appendix A. Supplementary data expression array data. Journal of Cellular Biochemistry. https://doi.org/10.1002/
jcb.10073. Article first published online.
Sun, Z., Wang, Y., Tan, T., Cui, J., 2005. Improving iris recognition accuracy via
Supplementary data related to this article can be found at cascaded classifiers. IEEE Trans. Syst. Man Cybern C 35 (3), 435e441.
https://doi.org/10.1016/j.diin.2018.03.002. Tan, Tieniu, He, Zhaofeng, Sun, Zhenan, 2010. Efficient and robust segmentation of
noisy iris images for non-cooperative iris recognition. Image Vis Comput. 28
(No. 2), pp. 223230.
References Thalheim, Lisa, Krissler, Jan, Ziegler, Peter-Michael, 2002. Body check: Biometric
access protection devices and their programs put to the test, vol. 11/2002,
Courtois, Nicolas, Pieprzyk, Josef, 2002. Cryptanalysis of block ciphers with over pp. 114e117. Available online: http://www.heise.de/ct/english/02/11/114/.
defined systems of equations. ASIACRYPT 267e287. Tobias, Orlando J., Seara, Rui, DECEMBER 2002. Image segmentation by histogram
Daugman, J., 2003. The importance of being random: statistical principles of iris thresholding using fuzzy sets. IEEE Trans. Image Process. 11 (12), 1457.
recognition. Pattern Recogn. 36 (2), 279e291. Wei, Z., Tan, T., Sun, Z., 2007. Nonlinear Iris deformation correction based on
Daugman, J., 2003. Demodulation by complex-valued wavelets for stochastic Gaussian model. Int. Conf. Biometrics 780e789.
pattern recognition. Int. J. Wavelets, Multiresolut. Inf. Process. 1 (1), 1e17. Wildes, R.P., 1997. Iris recognition: an emerging biometric technology. Proc. IEEE 85
Daugman, J., 2004. How iris recognition works. IEEE Trans. Circ. Syst. Video Technol. (9), 1348e1363.
14 (1), 21e30. Zhang, D., 2000. Automated Biometrics: Technologies and Systems. Kluwer,
Daugman, J., 2004. Iris recognition and anti-spoofing countermeasures. In: The 7th Norwell.
International Biometrics Conference, London.
104 V.K. Sinha et al. / Digital Investigation 25 (2018) 97e104
Vijay Kumar Sinha received his B.Tech.Degree in CSE from Dr. Manish Mahajan is holding Ph.D. degree from IKGPTU
Thapar University, Patiala, Punjab(India).He Received in Computer science and Engineering. His area of research
M.Tech. degree from Punjabi University, Patiala, Punja- includes image processing, Steganography. He is currently
b(India). He is researching in the field of security working as Prof.& HOD at CGCCOE, Landran, Mohali,
enhancement of iris recognition system since few years as Punjab(India) affiliated under IKGPTU, Kapurthala.
Ph.D. research scholr of Punjab Technical University,
Kapurthala (Punjab) India. He is currently working as
Professor at Chandigarh Engineering College, Landran,
Mohali (Punjab) India. His research interests includes Iris
recognition, Biometric authentication, Image Processing,
Fake biometric detection etc