指纹识别算法研究
|
摘要
随着计算机和网络的迅速发展,人们对身份认证的准确性、安全性与实用性提出了更高的要求。基于生物特征识别的智能身份认证技术也逐渐受到广泛的关注。在众多的生物识别技术中,指纹识别技术是发展最早、应用最广泛的一种。指纹识别技术充分利用了指纹的普遍性、唯一性和永久性的生物特征,已逐步取代了传统的基于标志和数字的识别方式,目前在网络、银行、金融、医疗和安检等行业均得到了广泛应用。
本文对指纹识别系统的原理和基本过程进行了分析研究,重点研究了指纹图像预处理算法,并且进行了验证。
首先,针对传统模板滤波后指纹脊线粘连及各个方向过渡不均的情况,提出了一种对称滤波模版算法,该算法采用以圆形直径旋转的滤波模版,在指纹图像不同方向上滤波效果均等,并有效的矫正了脊线的方向,去除了图像模糊引起的脊线边缘粘连,得到脊线和谷线了分离且平滑过渡滤波图像。
再次,详细研究了改进OPTA算法,针对该算法引起的毛刺较多及分叉点处细化不全问题进行了详细的研究,改进了消除模板,有效地去除水平方向的毛刺,提出了一种可变的消除模版和矫正消除模版,用可变的消除模版消除了垂直方向的毛刺,用矫正消除模版,使分叉点处细化更彻底,最终得到了比较光滑的细化图像。
最后,研究了指纹识别过程中伪特征的滤除方法,针对细节点提取过程中,图像边缘点作为特征点提取出来的这一问题,提出了一种边缘信息去伪法,有效的消除了边缘特征点。
测试图像的直观显示证明了这些研究方法的有效性。
With the rapid Progress of computer science and network technique,An accurate,secure and practical techno1ogy of Personal identification becomes more and more important. Technology of Personal identification based on Biometrics has received extensive attention. Technology of Fingerprint recognition is the earliest one and is app1ied widely in the all techniques of biometrics recognition, Then is taking full advantage of the universality, uniqueness and permanency of the fingerprint, and gradually has taken place of traditional identification method that is based on symbol and number. Nowadays the technology of fingerprint identification is fully used in network, bank, finance, insurance and security.
This paper research the basic principles and process of the fingerprint identification system,and focus on the pre-processing algorithms of fingerprint image and finally verify.
Firstly, aiming at conglutination of ridge line and inhomogeneous transition in all directions in the traditional filter template, then proposed symmetrical filter template algorithm ,the method used rotate the filter template of circular diameter to equal filtering effect in different directions, corrected the orientation of ridge line, removed conglutination of ridge line, and separated the ridge from the valley clearly.
Secondly, this paper study deeply improved opta algorithm,aiming at incomplete thinning and many burrs, used improved elimination of templates, removed burrs of the horizontal direction effectively; proposed changeable template and correcting template, the changeable template eliminate the vertical direction of the burr, correcting template made the bifurcation point thoroughly and eventually figured out smooth fingerprint image .
Finally, this paper study elimination of false minutiae in process of the fingerprint identification, aiming at extracting border points as the feature points in the process of extracting minutiae, this paper proposed judge of border information and eliminated effectively border minutiae.
The test images prove the validity of the methods.
本文对指纹识别系统的原理和基本过程进行了分析研究,重点研究了指纹图像预处理算法,并且进行了验证。
首先,针对传统模板滤波后指纹脊线粘连及各个方向过渡不均的情况,提出了一种对称滤波模版算法,该算法采用以圆形直径旋转的滤波模版,在指纹图像不同方向上滤波效果均等,并有效的矫正了脊线的方向,去除了图像模糊引起的脊线边缘粘连,得到脊线和谷线了分离且平滑过渡滤波图像。
再次,详细研究了改进OPTA算法,针对该算法引起的毛刺较多及分叉点处细化不全问题进行了详细的研究,改进了消除模板,有效地去除水平方向的毛刺,提出了一种可变的消除模版和矫正消除模版,用可变的消除模版消除了垂直方向的毛刺,用矫正消除模版,使分叉点处细化更彻底,最终得到了比较光滑的细化图像。
最后,研究了指纹识别过程中伪特征的滤除方法,针对细节点提取过程中,图像边缘点作为特征点提取出来的这一问题,提出了一种边缘信息去伪法,有效的消除了边缘特征点。
测试图像的直观显示证明了这些研究方法的有效性。
With the rapid Progress of computer science and network technique,An accurate,secure and practical techno1ogy of Personal identification becomes more and more important. Technology of Personal identification based on Biometrics has received extensive attention. Technology of Fingerprint recognition is the earliest one and is app1ied widely in the all techniques of biometrics recognition, Then is taking full advantage of the universality, uniqueness and permanency of the fingerprint, and gradually has taken place of traditional identification method that is based on symbol and number. Nowadays the technology of fingerprint identification is fully used in network, bank, finance, insurance and security.
This paper research the basic principles and process of the fingerprint identification system,and focus on the pre-processing algorithms of fingerprint image and finally verify.
Firstly, aiming at conglutination of ridge line and inhomogeneous transition in all directions in the traditional filter template, then proposed symmetrical filter template algorithm ,the method used rotate the filter template of circular diameter to equal filtering effect in different directions, corrected the orientation of ridge line, removed conglutination of ridge line, and separated the ridge from the valley clearly.
Secondly, this paper study deeply improved opta algorithm,aiming at incomplete thinning and many burrs, used improved elimination of templates, removed burrs of the horizontal direction effectively; proposed changeable template and correcting template, the changeable template eliminate the vertical direction of the burr, correcting template made the bifurcation point thoroughly and eventually figured out smooth fingerprint image .
Finally, this paper study elimination of false minutiae in process of the fingerprint identification, aiming at extracting border points as the feature points in the process of extracting minutiae, this paper proposed judge of border information and eliminated effectively border minutiae.
The test images prove the validity of the methods.
引文
[1]边肇棋等编著.模式识别.北京:清华大学出版社,1990,20-27
[2]汤若冰.指纹识别的历史及与其它生物识别的比较.中国防伪报道.2006,47-50
[3]田捷,杨鑫.生物特征识别技术理论与应用.北京.电子工业出版社,2006,126-129
[4] Henry Fuzee, Fingerprint recognition in low quality images. Pattern Recognition, 1993,26(10):1441-1460
[5]甘露.指纹识别系统的研究与设计.武汉理工大学硕士学位论文.2006.6
[6]赵磊.自动指纹识别系统算法的研究.中南大学硕士学位论文.2007.5
[7] D.Maltoni,D.Maio,A,K,Jain,S.Prabhakar. Handbook of Fingerprint Recognition. Sprin -ger,New York,2003,3-16
[8]王波涛,孙景鳌,蔡安妮.指纹取像技术.北方工业大学学报,2001,13(1):17-20
[9]陈为民.指纹图像处理及匹配方法的研究.重庆大学博士学位论文.2006,4
[10]卢朝阳,张岗山,刘琳.指纹识别系统性能评价方法.西安电子科技大学学报(自然科学版),2002,29(6):804-808
[11] Salil Prabhakar. Fingerprint Classification and Matching Using a Filter bank: [Ph.D] US A:Michigan State University, 2001
[12]罗西平,田捷.自动指纹识别中的图像增强和细节匹配算法[J].软件学报,2002, 13(5):946-956
[13] Mehtre B M,Chatter J. Segmentation of fingerprint image a composite method, Pattern Recongnition. 1989,1(4):381-3855
[14] Rao C,Black K.Type classification of fingerprints:a syntactic approach[J].IEEE Tram.Pattern Analysis and Machine Intelligence,1980,2(3):223-231
[15] Jain A,Prabhakar S,Hong L. Filterbank-based fingerprint matching[J].IEEE Trans on image Processing,200,9(5):847-856
[16] Jain A,Hong L,Bolle R. On-line fingerprint verification[J].IEEE Trans on Pattern Analysis and Machine Intelligence, 1997,19(4):302-313
[17] MICHAEL M.S.CHONG, TAN HAN NGEE,LIU JUN. Geometric framework for Fingerprint image classification[J]. Pattern Recognition, 1997,30(9):1475-1488
[18] KAWAGIE M,TOJO A. Fingerprint pattern classification[J]. Pattern Recognition,1984,17(3):295-303
[19] MARR D , HILDRETH E C. Theory of edge detection [A]. Proc. R. Soc.London[C]. 1980,B207:1872-217
[20]马笑潇,黄席樾,周欣等.指纹自动识别系统中的关键技术—方向图[J].重庆大学学报(自然科学版), 2001,24(3):91-94.
[21]聂桂军,吴陈.自动指纹识别系统中连续分布方向图的研究[J].中国图像图形学报,2005,10(3):315-319
[22]聂桂军,吴陈等.一种新的指纹方向图——连续分布方向图的研究[J].华东船舶工业学学报(自然科学版),2004,18(6):60-65
[23] Sherlock D,Monro D M, Milland K. Fingerprint enhancement by directional fourier filtering[J]. IEEE Proc. Visual Image Signal Processing,1994,141(2):87-94
[24] Hong L,Wan Y,Jain A K. Fingerprint image enhancement:algorithm and perform ance evaluation[J]. IEEE Trans.Pattern Analysis and Machine Intelligence, 1998, 20(8):777-789
[25]温苗利,梁彦等.基于Gabor函数的小波域指纹图像增强算法.计算机应用,2006
[26] Bovik A C, Clark M, Geisler W S.Multichannel texture analysis using localized spatial filters [J]. IEEE Trans Pattern Anal Machine Intel,1990,55(12):798-805
[27] O’Gorman L,Nickerson J V. An approach to fingerprint filter design[J].Patter Recongnition,1989,22(1):29-38
[28]张明,吴陈,陈楠.基于方向滤波的指纹图像增强算法[J].微机发展,2005,15(6):85-87
[29]林喜荣等.一种利用动态滤波模板进行指纹图像增强的算法[J].清华大学学报(自然科学版),200l:41(8):37-40
[30]叶四民,陈福祥.指纹图像预处理中的二值化技术[J].自动化与仪器仪表.2001,2:30-34
[31] Amengnal J C, Juan A, Perez J C, et al. Real-time minutiae extraction in fingerprint images[A]. Proc.of the 6th Int. Conf: On Image proceeding and Applications (IPA97)[EB/OL].http://www.cs.unr.edu/chris Smith Papers 2003
[32]楚亚蕴.一种结合方向信息的指纹图像二值化算法[J].中国图像图形学报,2006,15(6):855-860
[33]冈萨雷斯.数字图像处理[M].北京.电子工业出版社,2005.9:200-230
[34]傅文杰,苗夺谦等.利用改进的主图形算法提取指纹骨架[J].计算机应用,2006,26(7):1602-1604
[35]唐为方,陈旭等.两种常用指纹图像细化算法的改进[J].山东轻工业学院学报, 2005,19(1):25-30
[36] Lhin R T, Wan HK, Stover D L, et al. A one-pass thinning algorithm and its parallel implementation[J].Computer Vision Graphics Image Processing,1987,40(1):30-40
[37]冯星奎,李林艳,颜祖泉.一种新的指纹图像细化算法[J].中国图像图形学报, 1999,4(10):835-838
[38]王业琳,宁新宝,尹义龙.指纹图像细化算法的研究[J].南京大学学报,2003, 39(4):468-475
[39]刘怀泉,指纹识别系统研究.山东大学硕士学位论文.2005.6
[40] Dario Maio. Direct Gray Scale Minutiae Detection in Fingerprints[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,1997,(19)1:27-40
[41] Jiang, W.Y.Yau and W.Ser. Detecting fingerprint minutiae adaptive tracing the gray level ridge. Pattern Recognition,2001,34(11):999-1013
[42]冯国进,顾国华,张保民.指纹图像预处理与特征提取[J].计算机应用研究, 2004(5):183-185
[43] X.Sun and Z. Ai-Automatic feature extraction and recognition of fingerprint imagel.Proceed ofICSP96, 1996,1086-1089
[44] X.Sun and Z.Ai-Automatic feature extraction and recognition of fingerprint imagel. Proceed of ICSP96, 1996,1086-1089
[45] Xiao Q, Raafat H. Fingerprint image processing: A combined statistical and structural approach [J]. Pattern Recognition,1991,24(10):985-992
[46] Douglas Hung DC. Enhancement and Feature Purification of Fingerprint Images[J]. Pattern Recognition, 1993, 26(11):1661-1671
[47] Ratha N, Chen S, Jain A K. Adaptive flow orientation based feature extraction in fingerprint images[J]. Pattern Recognition, 1995, 28 (11):1657-1672
[48] FarinaA, Vajna Z, Leone A. Fingerprint Minutiae Extraction from SkeletonizedBinary Images[J]. Pattern Recognition,1999,32(10):877-889
[49] TicoM, Kuosmanen P. An Algorithm for Fingerprint Image Post processing [C]. Proc. of the 34th A silom ar Conference on Signals, Systems and Computers, 2000. 1735-1739
[50] Bian Z, ZhangD, ShuW. Knowledge Based Fingerprint Postprocessing[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2002,16(1):53-56
[51]王建永,郭成安.一种基于局部结构信息的指纹伪特征滤除算法[J].中国图像图形学报,2003,8(12):1467-1474
[52]尹义龙,宁新宝,张晓梅.改进的指纹细节特征提取算法[J].中国图像图形学报,2002,7(12):1302-1306
[53] Ranade S, Rosenfeld A. Point pattern matching by relaxation[J]. Pattern Recognition, 1980,12(5):269-275
[54] ballard D H.Generalized hough transform to detect arbitrary pattern[J]. IEEE Trans. Pattern Analysis and Machine Intelligence,1981,3(2):111-122
[55] Stockman G, Kopstein S, Benett S. Matching images to models for registration and object detection via clustering[J].IEEE Trans. Pattern Analysis and Machine Intelligence, 1982,4(3):229-241
[56] Chang S H, Cheng F H, Hsu W Het al. Fast algorithm for point pattern matching: Invariant to translations, rotations, and scale changes[J]. Pattern Recognition, 1997,30(2):321-339
[57] Sparrow M, Sparrow P. A topological approach to the matching of single fingerprints:Development of algorithms for use on rolled impressions[M]. U.S.A.Government Printing Office,Washington, D.C.: National Bureau of Standards Special Publication, 1985
[58] Hrechak A, Mchugh J.Automated fingerprint recognition using structural matching [J]. Pattern Recognition, 1990,23(8):893-900
[59] Isenor D, Zaky S. Fingerprint identification using graph matching[J]. Pattern Recognition, 1986,19(2):113-122
[60]漆远,田捷,邓翔.基于遗传算法的指纹图匹配算法及应用[J].软件学报,2000,11(4):488-493
[61] Starink J P P, Backer Eric. Finding point correspondence using simulated annealing [J]. Pattern Recognition,1995,28(2):231-240
[62] Hong Lin. Automatic personal identification using fingerprints[D] East Lansing: Michigan State University,1998
[2]汤若冰.指纹识别的历史及与其它生物识别的比较.中国防伪报道.2006,47-50
[3]田捷,杨鑫.生物特征识别技术理论与应用.北京.电子工业出版社,2006,126-129
[4] Henry Fuzee, Fingerprint recognition in low quality images. Pattern Recognition, 1993,26(10):1441-1460
[5]甘露.指纹识别系统的研究与设计.武汉理工大学硕士学位论文.2006.6
[6]赵磊.自动指纹识别系统算法的研究.中南大学硕士学位论文.2007.5
[7] D.Maltoni,D.Maio,A,K,Jain,S.Prabhakar. Handbook of Fingerprint Recognition. Sprin -ger,New York,2003,3-16
[8]王波涛,孙景鳌,蔡安妮.指纹取像技术.北方工业大学学报,2001,13(1):17-20
[9]陈为民.指纹图像处理及匹配方法的研究.重庆大学博士学位论文.2006,4
[10]卢朝阳,张岗山,刘琳.指纹识别系统性能评价方法.西安电子科技大学学报(自然科学版),2002,29(6):804-808
[11] Salil Prabhakar. Fingerprint Classification and Matching Using a Filter bank: [Ph.D] US A:Michigan State University, 2001
[12]罗西平,田捷.自动指纹识别中的图像增强和细节匹配算法[J].软件学报,2002, 13(5):946-956
[13] Mehtre B M,Chatter J. Segmentation of fingerprint image a composite method, Pattern Recongnition. 1989,1(4):381-3855
[14] Rao C,Black K.Type classification of fingerprints:a syntactic approach[J].IEEE Tram.Pattern Analysis and Machine Intelligence,1980,2(3):223-231
[15] Jain A,Prabhakar S,Hong L. Filterbank-based fingerprint matching[J].IEEE Trans on image Processing,200,9(5):847-856
[16] Jain A,Hong L,Bolle R. On-line fingerprint verification[J].IEEE Trans on Pattern Analysis and Machine Intelligence, 1997,19(4):302-313
[17] MICHAEL M.S.CHONG, TAN HAN NGEE,LIU JUN. Geometric framework for Fingerprint image classification[J]. Pattern Recognition, 1997,30(9):1475-1488
[18] KAWAGIE M,TOJO A. Fingerprint pattern classification[J]. Pattern Recognition,1984,17(3):295-303
[19] MARR D , HILDRETH E C. Theory of edge detection [A]. Proc. R. Soc.London[C]. 1980,B207:1872-217
[20]马笑潇,黄席樾,周欣等.指纹自动识别系统中的关键技术—方向图[J].重庆大学学报(自然科学版), 2001,24(3):91-94.
[21]聂桂军,吴陈.自动指纹识别系统中连续分布方向图的研究[J].中国图像图形学报,2005,10(3):315-319
[22]聂桂军,吴陈等.一种新的指纹方向图——连续分布方向图的研究[J].华东船舶工业学学报(自然科学版),2004,18(6):60-65
[23] Sherlock D,Monro D M, Milland K. Fingerprint enhancement by directional fourier filtering[J]. IEEE Proc. Visual Image Signal Processing,1994,141(2):87-94
[24] Hong L,Wan Y,Jain A K. Fingerprint image enhancement:algorithm and perform ance evaluation[J]. IEEE Trans.Pattern Analysis and Machine Intelligence, 1998, 20(8):777-789
[25]温苗利,梁彦等.基于Gabor函数的小波域指纹图像增强算法.计算机应用,2006
[26] Bovik A C, Clark M, Geisler W S.Multichannel texture analysis using localized spatial filters [J]. IEEE Trans Pattern Anal Machine Intel,1990,55(12):798-805
[27] O’Gorman L,Nickerson J V. An approach to fingerprint filter design[J].Patter Recongnition,1989,22(1):29-38
[28]张明,吴陈,陈楠.基于方向滤波的指纹图像增强算法[J].微机发展,2005,15(6):85-87
[29]林喜荣等.一种利用动态滤波模板进行指纹图像增强的算法[J].清华大学学报(自然科学版),200l:41(8):37-40
[30]叶四民,陈福祥.指纹图像预处理中的二值化技术[J].自动化与仪器仪表.2001,2:30-34
[31] Amengnal J C, Juan A, Perez J C, et al. Real-time minutiae extraction in fingerprint images[A]. Proc.of the 6th Int. Conf: On Image proceeding and Applications (IPA97)[EB/OL].http://www.cs.unr.edu/chris Smith Papers 2003
[32]楚亚蕴.一种结合方向信息的指纹图像二值化算法[J].中国图像图形学报,2006,15(6):855-860
[33]冈萨雷斯.数字图像处理[M].北京.电子工业出版社,2005.9:200-230
[34]傅文杰,苗夺谦等.利用改进的主图形算法提取指纹骨架[J].计算机应用,2006,26(7):1602-1604
[35]唐为方,陈旭等.两种常用指纹图像细化算法的改进[J].山东轻工业学院学报, 2005,19(1):25-30
[36] Lhin R T, Wan HK, Stover D L, et al. A one-pass thinning algorithm and its parallel implementation[J].Computer Vision Graphics Image Processing,1987,40(1):30-40
[37]冯星奎,李林艳,颜祖泉.一种新的指纹图像细化算法[J].中国图像图形学报, 1999,4(10):835-838
[38]王业琳,宁新宝,尹义龙.指纹图像细化算法的研究[J].南京大学学报,2003, 39(4):468-475
[39]刘怀泉,指纹识别系统研究.山东大学硕士学位论文.2005.6
[40] Dario Maio. Direct Gray Scale Minutiae Detection in Fingerprints[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,1997,(19)1:27-40
[41] Jiang, W.Y.Yau and W.Ser. Detecting fingerprint minutiae adaptive tracing the gray level ridge. Pattern Recognition,2001,34(11):999-1013
[42]冯国进,顾国华,张保民.指纹图像预处理与特征提取[J].计算机应用研究, 2004(5):183-185
[43] X.Sun and Z. Ai-Automatic feature extraction and recognition of fingerprint imagel.Proceed ofICSP96, 1996,1086-1089
[44] X.Sun and Z.Ai-Automatic feature extraction and recognition of fingerprint imagel. Proceed of ICSP96, 1996,1086-1089
[45] Xiao Q, Raafat H. Fingerprint image processing: A combined statistical and structural approach [J]. Pattern Recognition,1991,24(10):985-992
[46] Douglas Hung DC. Enhancement and Feature Purification of Fingerprint Images[J]. Pattern Recognition, 1993, 26(11):1661-1671
[47] Ratha N, Chen S, Jain A K. Adaptive flow orientation based feature extraction in fingerprint images[J]. Pattern Recognition, 1995, 28 (11):1657-1672
[48] FarinaA, Vajna Z, Leone A. Fingerprint Minutiae Extraction from SkeletonizedBinary Images[J]. Pattern Recognition,1999,32(10):877-889
[49] TicoM, Kuosmanen P. An Algorithm for Fingerprint Image Post processing [C]. Proc. of the 34th A silom ar Conference on Signals, Systems and Computers, 2000. 1735-1739
[50] Bian Z, ZhangD, ShuW. Knowledge Based Fingerprint Postprocessing[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2002,16(1):53-56
[51]王建永,郭成安.一种基于局部结构信息的指纹伪特征滤除算法[J].中国图像图形学报,2003,8(12):1467-1474
[52]尹义龙,宁新宝,张晓梅.改进的指纹细节特征提取算法[J].中国图像图形学报,2002,7(12):1302-1306
[53] Ranade S, Rosenfeld A. Point pattern matching by relaxation[J]. Pattern Recognition, 1980,12(5):269-275
[54] ballard D H.Generalized hough transform to detect arbitrary pattern[J]. IEEE Trans. Pattern Analysis and Machine Intelligence,1981,3(2):111-122
[55] Stockman G, Kopstein S, Benett S. Matching images to models for registration and object detection via clustering[J].IEEE Trans. Pattern Analysis and Machine Intelligence, 1982,4(3):229-241
[56] Chang S H, Cheng F H, Hsu W Het al. Fast algorithm for point pattern matching: Invariant to translations, rotations, and scale changes[J]. Pattern Recognition, 1997,30(2):321-339
[57] Sparrow M, Sparrow P. A topological approach to the matching of single fingerprints:Development of algorithms for use on rolled impressions[M]. U.S.A.Government Printing Office,Washington, D.C.: National Bureau of Standards Special Publication, 1985
[58] Hrechak A, Mchugh J.Automated fingerprint recognition using structural matching [J]. Pattern Recognition, 1990,23(8):893-900
[59] Isenor D, Zaky S. Fingerprint identification using graph matching[J]. Pattern Recognition, 1986,19(2):113-122
[60]漆远,田捷,邓翔.基于遗传算法的指纹图匹配算法及应用[J].软件学报,2000,11(4):488-493
[61] Starink J P P, Backer Eric. Finding point correspondence using simulated annealing [J]. Pattern Recognition,1995,28(2):231-240
[62] Hong Lin. Automatic personal identification using fingerprints[D] East Lansing: Michigan State University,1998