虹膜识别算法的研究和实现
|
摘要
生物识别技术是近几年来热门研究课题之一，该技术是一门多学科交叉的技术，包括生物学、医学、计算机科学、电子学等。虹膜识别技术是其中极有潜力的生物识别技术，由于其具有不须接触、虹膜个体差异大、形态稳定、不易篡改等特性，在生物识别领域被称为“光学指纹”，具有巨大的应用前景。本文对虹膜识别算法进行了深入研究，期望能拥有自己的专利算法，服务于国民经济的各部门，如保密机关、银行自动存取款系统及各种需要身份校验的地方等。
在本文中，简单介绍了几种典型的生物识别技术的发展、详细地阐述了虹膜识别技术的发展及研究意义、虹膜生理结构及虹膜识别系统的组成。较深入地研究了虹膜识别算法，并实现了其各个步骤，即定位、预处理、特征提取和匹配。本文的研究工作主要集中在对定位、预处理、特征提取的研究。
在定位方面，提出了一种新的定位算法，即利用灰度特征均值与Hough变换相结合的定位算法，弥补了直接Hough变换方法和圆边缘检测算子的定位算法定位速度慢的问题。
在预处理方面，实现了虹膜由圆环到矩形的拉伸变换，使虹膜识别具有伸缩和平移不变性。同时利用直方图均衡化的方法消除了由于光照不均所造成的影响。
在虹膜纹理的特征提取方面，提出了采用Gabor变换的特性对虹膜纹理进行编码的两种方法，即离散变换编码的方法和频相编码的方法，并且详细地研究了选取不同子图像对虹膜识别结果的影响。同时利用海明距离进行模式匹配及结果分析工作，取得了良好的识别结果。
最后，将本文的成果及结论加以总结，并对今后的研究工作提出自己的几点期望。
Biometric recognition technique is one of the hottest research fields, which is the multi-discipline technology, including biology, anatomy, computer, electronics, etc.. Iris recognition technique is potential because it is untouched, the texture of iris is different from each other, morphology of iris is stable, and iris is difficult to modify etc.. It has been called "Optical Fingerprint" in the field of biometric recognition. So it has enormous application prospects. In the paper, the iris recognition algorithm is researched deeply in order to own our patent that can serve for the departments such as secret department, bank automatic system and other domains that need identification.
In this paper, the development of biometric recognition and some kinds of biometric recognition techniques are introduced. Then the development of iris recognition technique and the structure of iris recognition system are discussed in detail. The algorithm of iris recognition has also been researched deeply and all of the processes that include location, pre-processing, feature extraction, and pattern match have been carried out Location and feature extraction have been focused on researching since they are the key steps of them.
In the iris location respect, a new location method is put forward which can enhance the speed of location. Then a mapped method is used to achieve the coordinate transform. For feature extraction, two kinds of algorithm based on Gabor transform are advanced. One of them is discrete Gabor encoding method and the other is frequency and phase encoding method. And the selection of child image that effects the results of iris recognition has been studied in detail. Then the matching algorithms are applied and he consequences of recognition are very good.
At last the results and conclusions of the paper are pointed out. At the same time, some expectations to future research are predicted.
在本文中，简单介绍了几种典型的生物识别技术的发展、详细地阐述了虹膜识别技术的发展及研究意义、虹膜生理结构及虹膜识别系统的组成。较深入地研究了虹膜识别算法，并实现了其各个步骤，即定位、预处理、特征提取和匹配。本文的研究工作主要集中在对定位、预处理、特征提取的研究。
在定位方面，提出了一种新的定位算法，即利用灰度特征均值与Hough变换相结合的定位算法，弥补了直接Hough变换方法和圆边缘检测算子的定位算法定位速度慢的问题。
在预处理方面，实现了虹膜由圆环到矩形的拉伸变换，使虹膜识别具有伸缩和平移不变性。同时利用直方图均衡化的方法消除了由于光照不均所造成的影响。
在虹膜纹理的特征提取方面，提出了采用Gabor变换的特性对虹膜纹理进行编码的两种方法，即离散变换编码的方法和频相编码的方法，并且详细地研究了选取不同子图像对虹膜识别结果的影响。同时利用海明距离进行模式匹配及结果分析工作，取得了良好的识别结果。
最后，将本文的成果及结论加以总结，并对今后的研究工作提出自己的几点期望。
Biometric recognition technique is one of the hottest research fields, which is the multi-discipline technology, including biology, anatomy, computer, electronics, etc.. Iris recognition technique is potential because it is untouched, the texture of iris is different from each other, morphology of iris is stable, and iris is difficult to modify etc.. It has been called "Optical Fingerprint" in the field of biometric recognition. So it has enormous application prospects. In the paper, the iris recognition algorithm is researched deeply in order to own our patent that can serve for the departments such as secret department, bank automatic system and other domains that need identification.
In this paper, the development of biometric recognition and some kinds of biometric recognition techniques are introduced. Then the development of iris recognition technique and the structure of iris recognition system are discussed in detail. The algorithm of iris recognition has also been researched deeply and all of the processes that include location, pre-processing, feature extraction, and pattern match have been carried out Location and feature extraction have been focused on researching since they are the key steps of them.
In the iris location respect, a new location method is put forward which can enhance the speed of location. Then a mapped method is used to achieve the coordinate transform. For feature extraction, two kinds of algorithm based on Gabor transform are advanced. One of them is discrete Gabor encoding method and the other is frequency and phase encoding method. And the selection of child image that effects the results of iris recognition has been studied in detail. Then the matching algorithms are applied and he consequences of recognition are very good.
At last the results and conclusions of the paper are pointed out. At the same time, some expectations to future research are predicted.
引文
[1] Jain A K, Bolle R and Pankanti S. Biometrics Personal Identification in Networked Society. Kluwer Academic Publishers, 1999.
[2] Clarke R. Human Identification in Information System: Management Challenges and Public Policy Issues. Information Technology&People, 1994, 7(4): 6～37
[3] Flom L and Safir A. Iris Recognition System. US: Patent 4641349, 1987.
[4] Daugman J G.. Biometrical Personal Identification System Based on Iris Analysis. US: Patent 52911560, 1994.
[5] Wildes R P. Automated, Non-invasive Iris Recognition System and Method. US:Patent 5572596, 1996.
[6] Boles W W. A Secutity System Based on Hunam Iris Identification Using Wavelet Transform. IEEE, 1997 First International Conference on Knowledge-based Intelligent Electronic Systems. 1997: 533～541
[7] 谭铁牛，朱勇，王蕴红．基于虹膜识别的身份鉴别方法与装置．发明专利．国家知识产权局专利申请号：99110825．6．1999．
[8] 康浩．虹膜识别系统研究：学位论文．上海：上海交通大学，1999
[9] 应忍冬．虹膜身份识别方法的研究：学位论文．上海：上海交通大学，2000
[10] 陈良洲，叶虎年．一种新的虹膜识别算法研究．华北工学院测试技术学报，2000，14(4)：211～216
[11] 严民军，汪云九．虹膜识别的计算机识别原理．生物化学与生物物理进展，2000，27(4)．348～350
[12] Wildes R P. Iris recognition:an emerging biometric technology. Proceding of IEEE, 1997, 85 (9): 1348-1363
[13] 朱勇．基于生物特征的身份鉴别研究：学位论文．北京：中科院自动化研究所，2000
[14] http://www.cmi.hku.hk/Teaching/eye/structure/structure.htm
[15] Adler F H..Physiology of the Eye. Mosby St.Louis: MO, 1965.
[16] Kronfeld P C. The Gross Anatomy and the Embryology of the Eye. The Eye, 1986(1): 1-66
[17] Mann I. The Developmnet of the Human Eye. New York: Grime and Stratton, 1950.
[18] 贾云德．机器视觉．北京：科学出版社，2000．97～100
[19] Canny J. A. Computional Approach to Edge detection. IEEE Trans. Patern Analysis and Machine Intelligence, 1986, 8(6): 679～698
[20] 阮秋琦．数字图像处理学．北京：电子工业出版社，2001．122-178，417～418，419～425
[21] Hough P V C. Method and means for recognizing complex patterns. US: Patent 3069654, 1962
[22] 容观澳．计算机图像处理．北京：清华大学出版社，2000．2．257～259
[23] 章毓晋．图像处理和分析．北京：清华大学出版社．1999．3．181～192。
[24] 何家峰，叶虎年．虹膜定位．中国图像图形学报 A，2000，5(3)：253～255．
[25] 韩方，陈颖，陆亭立．虹膜定位算法．上海大学学报(自然科学版)，2001，7(6)：500～502
[26] Daugman J G. High Confidence Visual Recognition of Persons by a Test of Statistical Independence. IEEE Trans. Patern Analysis and Machine Intelligence, 1993, 15: 1148～1161
[27] Boles W and Boashah B. A Huaman Identification Technique Using Images of the Iris and Wavelet Transform. IEEE Trans. on Signal Processing, 1998, 46: 1185～1188
[28] Wildes R P. A System for automated iris recogntion. Proc. 2nd IEEE Workshop Applicat. Comput .Vision, Dec 1994, 121～128
[29] Wildes R P. Automated, Non-invasive Iris Recognition System and Method. US: Patent 5572596, 1996
[30] 徐涛．数值计算方法．吉林科学技术出版社．1998．7
[31] Shinyoung Lim, Kwanyong Lee, Okhwan Byeon, and Taiyun Kim. Efficient iris recogntion through Improvement of feature vector and classifier. June 2001, ETRI Journal, 23(6), 62～70
[32] Williams G O. Iris Recogntion Technology. IEEE AES System Magazine, April 1997, 23～39。
[33] Daugman J G and Downing C. Recognizing Iris Texture by Phase Demodulation. The Institution of Electrical Engineers. IEE, 1994, 1～8
[34] Turner M R. Texture discrimination by Gabor function. Biol Cybern, 1986, 55: 71～82
[35] Clark M, Borik A C, and Geisler W S. Texture segmentation using a class of narrow band filters, Proc. IEEE. Trran on Pattern Analysis &Mancine intelligence, 1990.
[36] Gabor D. Theory of Communication. J. lnst. Elect, Eng, 1946, 93: 429～457
[37] Jones J and Palmer L, An evalutation of the two-dimensinal Gabor filter model of simple receptive fields in cat striate cortex. J.Neurophysiol, 1987, 58: 1233～1258
[38] Daugman J G. Complete Discrete 2_D Gabor Transforms by Neural Networks of image Analysis and Compression. IEEE, Trans.onAcoustics, Speech&SignalProcessing, 1988, 36(7): 1169～1179
[39] Daugman J G. Uncertainty Relation For Resolutin in Space, Spatial Frequency and Orientation Optimized By Two-Dimensional Visual Cortical filters. 1985, 12(7): 1160～1169
[40] http://www.iridiantech.com/products.php?page=4
[41] http://www.evermedia.co.kr/eng/main.php?viewPage=goods
[42] Haralick R M. Statistical and structural approaches to texture. Proc, IEEE, 1979, 67: 786～804
[43] Jain J and Farkokhnia F. Unsupervised texture segmentation using gabor filters. Pattern Recognition, 1991, 24(2): 1167～1185
[44] Bastiaans M. Gabor's expansion of a signal into Gaussian elementary signal. Proc IEEE, 1980, 68: 538～539
[45] Marcelja S .Mathematical descrition of the responses of simple cortical cells, J.Opt.Soc. Amer, 1980, 70: 1297～1300
[46] Samal A and Iyengar P A. Automatic recogntion and analysi of human faces and facial expressions:Asurvey. Pattern Recognit, 1992, 25: 65～77.
[47] Mallat S G. A theory for multiresolution signal decompostion:The wavelet representation. IEEE Trans Pattern Anal.Machine Intell, 1989, 1: 674～693
[48] Logan B F. Information in the zero-crossings of bandpass signals. Bell Syst.Tech.J., 1977, 56: 487～501
[49] Pratt W K. Digital Image Processing. New York: Lwiley, 1978.
[50] Castleman K R. Digital Image Processing. Prentice-Hall Internatioanl, Inc. 1995, 426～429
[51] 康浩，徐国治．虹膜纹理像为编码．上海交通大学学报，1999，33(1)：542～544
[52] 康浩，徐国治．虹膜识别系统．电路与系统学报，2000，5(1)：11～15
[2] Clarke R. Human Identification in Information System: Management Challenges and Public Policy Issues. Information Technology&People, 1994, 7(4): 6～37
[3] Flom L and Safir A. Iris Recognition System. US: Patent 4641349, 1987.
[4] Daugman J G.. Biometrical Personal Identification System Based on Iris Analysis. US: Patent 52911560, 1994.
[5] Wildes R P. Automated, Non-invasive Iris Recognition System and Method. US:Patent 5572596, 1996.
[6] Boles W W. A Secutity System Based on Hunam Iris Identification Using Wavelet Transform. IEEE, 1997 First International Conference on Knowledge-based Intelligent Electronic Systems. 1997: 533～541
[7] 谭铁牛，朱勇，王蕴红．基于虹膜识别的身份鉴别方法与装置．发明专利．国家知识产权局专利申请号：99110825．6．1999．
[8] 康浩．虹膜识别系统研究：学位论文．上海：上海交通大学，1999
[9] 应忍冬．虹膜身份识别方法的研究：学位论文．上海：上海交通大学，2000
[10] 陈良洲，叶虎年．一种新的虹膜识别算法研究．华北工学院测试技术学报，2000，14(4)：211～216
[11] 严民军，汪云九．虹膜识别的计算机识别原理．生物化学与生物物理进展，2000，27(4)．348～350
[12] Wildes R P. Iris recognition:an emerging biometric technology. Proceding of IEEE, 1997, 85 (9): 1348-1363
[13] 朱勇．基于生物特征的身份鉴别研究：学位论文．北京：中科院自动化研究所，2000
[14] http://www.cmi.hku.hk/Teaching/eye/structure/structure.htm
[15] Adler F H..Physiology of the Eye. Mosby St.Louis: MO, 1965.
[16] Kronfeld P C. The Gross Anatomy and the Embryology of the Eye. The Eye, 1986(1): 1-66
[17] Mann I. The Developmnet of the Human Eye. New York: Grime and Stratton, 1950.
[18] 贾云德．机器视觉．北京：科学出版社，2000．97～100
[19] Canny J. A. Computional Approach to Edge detection. IEEE Trans. Patern Analysis and Machine Intelligence, 1986, 8(6): 679～698
[20] 阮秋琦．数字图像处理学．北京：电子工业出版社，2001．122-178，417～418，419～425
[21] Hough P V C. Method and means for recognizing complex patterns. US: Patent 3069654, 1962
[22] 容观澳．计算机图像处理．北京：清华大学出版社，2000．2．257～259
[23] 章毓晋．图像处理和分析．北京：清华大学出版社．1999．3．181～192。
[24] 何家峰，叶虎年．虹膜定位．中国图像图形学报 A，2000，5(3)：253～255．
[25] 韩方，陈颖，陆亭立．虹膜定位算法．上海大学学报(自然科学版)，2001，7(6)：500～502
[26] Daugman J G. High Confidence Visual Recognition of Persons by a Test of Statistical Independence. IEEE Trans. Patern Analysis and Machine Intelligence, 1993, 15: 1148～1161
[27] Boles W and Boashah B. A Huaman Identification Technique Using Images of the Iris and Wavelet Transform. IEEE Trans. on Signal Processing, 1998, 46: 1185～1188
[28] Wildes R P. A System for automated iris recogntion. Proc. 2nd IEEE Workshop Applicat. Comput .Vision, Dec 1994, 121～128
[29] Wildes R P. Automated, Non-invasive Iris Recognition System and Method. US: Patent 5572596, 1996
[30] 徐涛．数值计算方法．吉林科学技术出版社．1998．7
[31] Shinyoung Lim, Kwanyong Lee, Okhwan Byeon, and Taiyun Kim. Efficient iris recogntion through Improvement of feature vector and classifier. June 2001, ETRI Journal, 23(6), 62～70
[32] Williams G O. Iris Recogntion Technology. IEEE AES System Magazine, April 1997, 23～39。
[33] Daugman J G and Downing C. Recognizing Iris Texture by Phase Demodulation. The Institution of Electrical Engineers. IEE, 1994, 1～8
[34] Turner M R. Texture discrimination by Gabor function. Biol Cybern, 1986, 55: 71～82
[35] Clark M, Borik A C, and Geisler W S. Texture segmentation using a class of narrow band filters, Proc. IEEE. Trran on Pattern Analysis &Mancine intelligence, 1990.
[36] Gabor D. Theory of Communication. J. lnst. Elect, Eng, 1946, 93: 429～457
[37] Jones J and Palmer L, An evalutation of the two-dimensinal Gabor filter model of simple receptive fields in cat striate cortex. J.Neurophysiol, 1987, 58: 1233～1258
[38] Daugman J G. Complete Discrete 2_D Gabor Transforms by Neural Networks of image Analysis and Compression. IEEE, Trans.onAcoustics, Speech&SignalProcessing, 1988, 36(7): 1169～1179
[39] Daugman J G. Uncertainty Relation For Resolutin in Space, Spatial Frequency and Orientation Optimized By Two-Dimensional Visual Cortical filters. 1985, 12(7): 1160～1169
[40] http://www.iridiantech.com/products.php?page=4
[41] http://www.evermedia.co.kr/eng/main.php?viewPage=goods
[42] Haralick R M. Statistical and structural approaches to texture. Proc, IEEE, 1979, 67: 786～804
[43] Jain J and Farkokhnia F. Unsupervised texture segmentation using gabor filters. Pattern Recognition, 1991, 24(2): 1167～1185
[44] Bastiaans M. Gabor's expansion of a signal into Gaussian elementary signal. Proc IEEE, 1980, 68: 538～539
[45] Marcelja S .Mathematical descrition of the responses of simple cortical cells, J.Opt.Soc. Amer, 1980, 70: 1297～1300
[46] Samal A and Iyengar P A. Automatic recogntion and analysi of human faces and facial expressions:Asurvey. Pattern Recognit, 1992, 25: 65～77.
[47] Mallat S G. A theory for multiresolution signal decompostion:The wavelet representation. IEEE Trans Pattern Anal.Machine Intell, 1989, 1: 674～693
[48] Logan B F. Information in the zero-crossings of bandpass signals. Bell Syst.Tech.J., 1977, 56: 487～501
[49] Pratt W K. Digital Image Processing. New York: Lwiley, 1978.
[50] Castleman K R. Digital Image Processing. Prentice-Hall Internatioanl, Inc. 1995, 426～429
[51] 康浩，徐国治．虹膜纹理像为编码．上海交通大学学报，1999，33(1)：542～544
[52] 康浩，徐国治．虹膜识别系统．电路与系统学报，2000，5(1)：11～15