AN IMPROVED IRIS RECOGNITION SYSTEM BASED ON 2-D DCT AND HAMMING DISTANCE TECHNIQUE

  • 41 views
Uploaded on

This paper proposes a new iris recognition system that implements Integro-Differential, Daugman Rubber Sheet Model, 2-D DCT, Hamming Distance to exact features from the iris and matching it with the …

This paper proposes a new iris recognition system that implements Integro-Differential, Daugman Rubber Sheet Model, 2-D DCT, Hamming Distance to exact features from the iris and matching it with the sorted database.All these image-processing algorithms have been validated on noised real iris images & UBIRIS database

More in: Engineering
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
41
On Slideshare
0
From Embeds
0
Number of Embeds
1

Actions

Shares
Downloads
0
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. ICRTEDC-2014 32 Vol. 1, Spl. Issue 2 (May, 2014) e-ISSN: 1694-2310 | p-ISSN: 1694-2426 GV/ICRTEDC/08 AN IMPROVED IRIS RECOGNITION SYSTEM BASED ON 2-D DCT AND HAMMING DISTANCE TECHNIQUE Sakshi Sharma Electronics & Communication Engineering Department, Chandigarh Engineering College, Mohali, Punjab cecm.ece.ssh@gmail.com Abstract—The biometric person authentication technique based onthe pattern of the human iris is well suited to be applied to any access control system requiring a high level ofsecurity. This paper proposes a new iris recognition system that implements Integro-Differential, Daugman Rubber Sheet Model, 2-D DCT, Hamming Distance to exact features from the iris and matching it with the sotred database.All these image-processing algorithms have been validated on noised real iris images & UBIRISdatabase. The proposed innovative technique is computationally effective as well as reliable in terms of recognition rates. 1. INTRODUCTION Biometrics refers to the quantifiable data (or metrics) related to human characteristics and traits. Biometrics identification (or biometric authentication) is used in computer science as a form of identification and access control. It is also used to identify individuals in groups that are under surveillance. Biometric identifiers are the distinctive, measurable characteristics used to label and describe individuals. Biometric identifiers are often categorized as physiological versus behavioral characteristics. Physiological characteristics are related to the shape of the body. Examples include, but are not limited to fingerprint, face recognition, DNA, palm print, hand geometry, iris recognition, retina and odour/scent. Behavioral characteristics are related to the pattern of behavior of a person, including but not limited to typing rhythm, gait, and voice. Some researchers have coined the term behaviometrics to describe the latter class of biometrics. Types of Biometrics Biometric system is broadly categorized in two types: Physiological and behavioral. Figure 1: Types of biometrics I. Working principle of biometrics Biometrics device consists of a scanning device and software, that converts the gathered information into digital form, and a database or memory that stores the biometric data for comparison with previous records saved in the system. After converting the biometric input into digital form, the software identifies the match points in the data values. The match points are processed using algorithm into a value that can be compared with biometric data already stored in the data base. All biometric systems require comparing a registered biometric sample against a newly captured biometric sample. Advantages of Using Biometrics: Easier fraud detection. Better than password/PIN or smart cards. No need to memorize passwords. Require physical presence of the person to be identified. Physical characteristics are unique. It providesaccurate results. 2. BACKGROUND The below table shows the related research work: RESEARCHE R NAME YEA R ALGORITH M USED DRAWBACK S John G. Daugman 1994 Integro- Differential, Daugman Rubber Sheet Model, 2-D Gabor Filter, XOR operator Hamming Distance. Integro- differential operator fails in case of noise and total execution time is also very high. W. W. Boles and B. Boashash 1998 1-D wavelet transforms, Edge detection technique, Zero crossing representation . Algorithms are tested on few number of Iris images, Correct recognition rate is 92%, Equal Error rate is 8.13%.
  • 2. 33 ICRTEDC -2014 Zhonghua Lin and Bibo Lu 2010 Morlet wavelet transforms Polar co- ordinate transform. Recognition rate is low of the system. Bimi Jain, Dr. M.K. Gupta and Prof. Jyoti Bharti 2012 Fast Fourier transform, Euclidean distance for matching. Algorithm tested only on 10 images, FAR and FRR are also not declared and Euclidean distance technique make computational slow. Mohd. T. Khan 2013 1-D Log Gabor filter, K- dimensional tree technique for matching. Search efficiency is decreased by large tree size and FAR, FRR, ERR are not mentioned in results. 3. PROPOSED APPROACH Figure 2: Proposed approach SEGMENTATION The color image is firstly converted into gray scale image; it means that the luminance of colored image is converted into gray shade. The first stage of iris recognition is to isolate the actual iris region in a digital eye image. The iris region, shown in Figure 4, can be approximated by two circles, first one is for the iris boundary region and second one is for the pupil boundary region. The eyelids and eyelashes cover the upper and lower parts of the iris region. Specular reflections can also occur within the iris region resulting into corrupting the iris pattern. Figure 3:Grayscale iris image NORMALIZATION Once the segmentation module has estimated the iris’s boundary, the normalization process will transform the circular iris region into another shape which will have the same constant dimensions [8]. We can be using Daugman’s Rubber Sheet Model for normalization. This model transforms the iris texture from Cartesian to polar coordinates. This process is called as iris unwrapping, which have a rectangular entity that is used for further subsequent processing. The transformation of normal Cartesian to polar coordinates is recommended which maps the entire pixels in the iris area into a pair of polar coordinates (r, θ), where r and θ represents the intervals of [0 1] and[0 2π] as shown in figure 5. Daugman’s Rubber Sheet Model: For normalization Daugman has invented the Rubber Sheet Model in which he remaps each point within the iris region to a pair of polar coordinates (r,θ) where r is on the interval [0,1] and θ is angle [0,2π]. Normalisation accounts for variations in pupil size due to changes in external illumination that might influence iris size, it also ensures that the irises of different individuals are mapped onto a common image domain in spite of the variations in pupil size across subjects etc. Figure 4: Normalized Iris FEATURE EXTRACTION After the iris is normalized, it is compressed by using mathematical functions and converted in to binary forms. Each isolated iris pattern is then encoded using DCT method to extract its binary information. Discrete Cosine Transform: A discrete cosine transform (DCT) expresses a finite sequence of data points in terms of a sum of cosine functions oscillating at different frequencies. DCT algorithm is very efficient in image compression applications which makes further computational easy in the system. Discrete cosine transform provides the output in the form of matrix. 33 ICRTEDC -2014 Zhonghua Lin and Bibo Lu 2010 Morlet wavelet transforms Polar co- ordinate transform. Recognition rate is low of the system. Bimi Jain, Dr. M.K. Gupta and Prof. Jyoti Bharti 2012 Fast Fourier transform, Euclidean distance for matching. Algorithm tested only on 10 images, FAR and FRR are also not declared and Euclidean distance technique make computational slow. Mohd. T. Khan 2013 1-D Log Gabor filter, K- dimensional tree technique for matching. Search efficiency is decreased by large tree size and FAR, FRR, ERR are not mentioned in results. 3. PROPOSED APPROACH Figure 2: Proposed approach SEGMENTATION The color image is firstly converted into gray scale image; it means that the luminance of colored image is converted into gray shade. The first stage of iris recognition is to isolate the actual iris region in a digital eye image. The iris region, shown in Figure 4, can be approximated by two circles, first one is for the iris boundary region and second one is for the pupil boundary region. The eyelids and eyelashes cover the upper and lower parts of the iris region. Specular reflections can also occur within the iris region resulting into corrupting the iris pattern. Figure 3:Grayscale iris image NORMALIZATION Once the segmentation module has estimated the iris’s boundary, the normalization process will transform the circular iris region into another shape which will have the same constant dimensions [8]. We can be using Daugman’s Rubber Sheet Model for normalization. This model transforms the iris texture from Cartesian to polar coordinates. This process is called as iris unwrapping, which have a rectangular entity that is used for further subsequent processing. The transformation of normal Cartesian to polar coordinates is recommended which maps the entire pixels in the iris area into a pair of polar coordinates (r, θ), where r and θ represents the intervals of [0 1] and[0 2π] as shown in figure 5. Daugman’s Rubber Sheet Model: For normalization Daugman has invented the Rubber Sheet Model in which he remaps each point within the iris region to a pair of polar coordinates (r,θ) where r is on the interval [0,1] and θ is angle [0,2π]. Normalisation accounts for variations in pupil size due to changes in external illumination that might influence iris size, it also ensures that the irises of different individuals are mapped onto a common image domain in spite of the variations in pupil size across subjects etc. Figure 4: Normalized Iris FEATURE EXTRACTION After the iris is normalized, it is compressed by using mathematical functions and converted in to binary forms. Each isolated iris pattern is then encoded using DCT method to extract its binary information. Discrete Cosine Transform: A discrete cosine transform (DCT) expresses a finite sequence of data points in terms of a sum of cosine functions oscillating at different frequencies. DCT algorithm is very efficient in image compression applications which makes further computational easy in the system. Discrete cosine transform provides the output in the form of matrix. 33 ICRTEDC -2014 Zhonghua Lin and Bibo Lu 2010 Morlet wavelet transforms Polar co- ordinate transform. Recognition rate is low of the system. Bimi Jain, Dr. M.K. Gupta and Prof. Jyoti Bharti 2012 Fast Fourier transform, Euclidean distance for matching. Algorithm tested only on 10 images, FAR and FRR are also not declared and Euclidean distance technique make computational slow. Mohd. T. Khan 2013 1-D Log Gabor filter, K- dimensional tree technique for matching. Search efficiency is decreased by large tree size and FAR, FRR, ERR are not mentioned in results. 3. PROPOSED APPROACH Figure 2: Proposed approach SEGMENTATION The color image is firstly converted into gray scale image; it means that the luminance of colored image is converted into gray shade. The first stage of iris recognition is to isolate the actual iris region in a digital eye image. The iris region, shown in Figure 4, can be approximated by two circles, first one is for the iris boundary region and second one is for the pupil boundary region. The eyelids and eyelashes cover the upper and lower parts of the iris region. Specular reflections can also occur within the iris region resulting into corrupting the iris pattern. Figure 3:Grayscale iris image NORMALIZATION Once the segmentation module has estimated the iris’s boundary, the normalization process will transform the circular iris region into another shape which will have the same constant dimensions [8]. We can be using Daugman’s Rubber Sheet Model for normalization. This model transforms the iris texture from Cartesian to polar coordinates. This process is called as iris unwrapping, which have a rectangular entity that is used for further subsequent processing. The transformation of normal Cartesian to polar coordinates is recommended which maps the entire pixels in the iris area into a pair of polar coordinates (r, θ), where r and θ represents the intervals of [0 1] and[0 2π] as shown in figure 5. Daugman’s Rubber Sheet Model: For normalization Daugman has invented the Rubber Sheet Model in which he remaps each point within the iris region to a pair of polar coordinates (r,θ) where r is on the interval [0,1] and θ is angle [0,2π]. Normalisation accounts for variations in pupil size due to changes in external illumination that might influence iris size, it also ensures that the irises of different individuals are mapped onto a common image domain in spite of the variations in pupil size across subjects etc. Figure 4: Normalized Iris FEATURE EXTRACTION After the iris is normalized, it is compressed by using mathematical functions and converted in to binary forms. Each isolated iris pattern is then encoded using DCT method to extract its binary information. Discrete Cosine Transform: A discrete cosine transform (DCT) expresses a finite sequence of data points in terms of a sum of cosine functions oscillating at different frequencies. DCT algorithm is very efficient in image compression applications which makes further computational easy in the system. Discrete cosine transform provides the output in the form of matrix.
  • 3. ICRTEDC-2014 34 MATCHING The matching algorithm consists of all the image processing steps that are carried out at the time of enrolling the encoded iris template in database. Once the bit encrypted bit pattern B’ corresponding to binary image formed is extracted, it is tried to match with all stored encrypted bit patterns B using simple Boolean XOR operation[2]. The dissimilarity measure between any two iris bit patterns is computed using Hamming Distance (HD) which is given as, Where, N=total number of bits in each bit pattern. As HD is a fractional measure of dissimilarity with 0 representing A perfect match, a low normalized HD implies strong similarity of iris codes. FIGURE 5: IRIS RECOGNITION TECHNOLOGY[21] 4. RESULTS & CONCLUSIONS This work proposes a modified iris recognition system based on 2D DCT and Daughman rubber sheet is used for normalization is based on minimizing the effect of the eyelids and eyelashes by trimming the iris area above the upper and the area below the lower boundaries of the pupil. The Experimental results also indicate that the performance of the proposed technique is computationally effective as well as reliable in terms of recognition rate of 93.2%. The combination of Daughman rubber sheet and 2D DCT is promising. REFERENCES [1] G K. Jain, L. Hong and S. Pankanti, Biometrics: Promising Frontiers for Emerging Identification Market, Comm. ACM ,pp. 91-98, Feb. 2000. [2] A. Ross, D. Nandakumar, A.K. Jain, Handbook of Multibiometrics, . Springer, Heidelberg (2006). [3] J. Daugman , How iris recognition works, IEEE Trans. onCircuits and Systems for Video Technology., Vol. 14, No. 1,pp. 21-30, January 2004. [4] L. Flom, A. Safir, Iris recognition system, US Patent 4641394, 1987. [5] K.W. Bowyer, K. Hollingsworth, P. J. Flynn, Image Understanding for Iris Biometrics: A Survey, Computer vision and Image Understanding, Vol. 110, Issue 2, pp. 281- 307, 2008. [ 6] J. Daugman, High Confidence Visual Recognition of Persons by a Test of Statistical Independence, IEEE Trans.on Pattern Analysis and Machine Intelligence, Vol. 15, No.11, pp.1148-1161, 1993. [7] J. Daugman, C.Downing, Epigenetic randomness, complexityand singularity of human iris patterns, Proc. R. Soc. Lond. B268, pp. 1737–1740, 2001. [8] J. Daugman , How iris recognition works, IEEE Trans. onCircuits and Systems for Video Technology., Vol. 14, No. 1,pp. 21-30, January 2004. [9] Center for Biometrics and Security Research, CASIA Iris ImageA. K Jain, P. Flynn, and A. Ross, Handbook of Biometrics, New York: Springer, 2008. [10] Sunita V. Dhavale ”DWT and DCT based Robust Iris Feature Extraction and Recognition Algorithm for Biometric Personal Identification”, International Journal of Computer Applications (0975 – 8887), Volume 40– No.7, February 2012. [11] J. Daugman “How iris recognition works”, Proceedings of 2002 International Conference on Image Processing, Vol.1, 2002. [12] J. Daugman. “Biometric personal identification system based on iris analysis” United States International Journal of Advanced Trends in Computer Science and Engineering, Vol.2, No.1, Pages : 93-97 (2013) Special Issue of ICACSE 2013 - Held on 7-8 January, 2013 in Lords Institute of Engineering and Technology, Hyderabad. [13] Human eye. “Encyclopedia Britannica” from Encyclopedia Britannica Ultimate Reference Suite DVD, 2006. [14] Libor Masek, “Recognition of Human Iris Patterns for Biometric Identification”, The University of Western Australia, 2003. [15] Bowyer K.W., Kranenburg C., Dougherty S. “Edge Detector Evaluation using Empirical ROC Curves”, IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp. 354-359,1999. [16]B.Sabarigiri1, T.Karthikeyan2, “Acquisition of Iris Images, Iris Localization, Normalization and Quality Enhancement for Personal Identification”, International Journal of Emerging Trends &Technology in Computer Science (IJETTCS),Volume 1, Issue 2, July – August 2012. [17] Jain, A., Hong, L., & Pankanti, S. (2000). "Biometric Identification". Communications of the ACM,43(2), p. 91- 98. DOI 10.1145/328236.328110. [18] Mayuri Memane, Sanjay Ganorkar, “DWT Based Iris Recognition”, International Journal of Engineering Science and Technology (IJEST),Vol. 4 No.08 August 2012. [19] John G. Daugman, (1994), “Biometric Personal Identification System based on Iris Analysis”, U.S. Patent no. 5291560 A, publish date 1 march, 1994. [20] Boles and Boashash, (1998), “A Human Identification Technique Using Images of the Iris and Wavelet Transform”, IEEE Transactions on Signal Processing, vol. 46, no. 4, pp. 1185-1188. [21] www.google.com