基于FPGA硬件方式实现的指纹识别系统研究
|
摘要
由于人体指纹具有唯一性、广泛性以及不可复制性等优点，因此指纹识别技术己发展成为应用最广的一种生物识别技术。随着人们对小型消费类设备，特别是移动设备的安全性要求的日渐提高，传统的自动指纹识别认证系统AFIS(Automatic Fingerprint Identification System)移动性较差、体积较大的弊端就显现出来了。因此研究应用在这些小型消费类产品中的AFIS具有很强的现实意义和广阔的市场空间。基于上述背景，本论文在深入研究并改进BST FingerAPI1.0指纹识别算法的基础上，选用可编程门阵列FPGA(Field Programmed Gate Array)来实现AFIS，主要做了以下几方面工作：
1、在整理学习大量基于DSP或ARM的指纹识别技术资料的基础上，对这类指纹识别技术特点、应用领域和发展趋势做了研究，并针对传统的嵌入式指纹识别系统的缺点提出了相应的解决方案。
2、深入调研了现有各种指纹识别算法，特别是应用在嵌入式AFIS的算法，并结合BST FingerAPI1.0版本算法提出了适用于嵌入FPGA硬件电路的指纹识别算法。该算法主要由指纹检测采集、图像增强、指纹特征点提取以及比对四个部分组成，与传统的算法相比，减少了指纹细化和二值化的过程。这种算法的改进在保证指纹识别系统性能的同时减少了FPGA资源占有量，提高了算法运算速度。
3、对基于FPGA的指纹识别系统进行结构和功能设计，把整个AFIS划分为指纹采集、算法、存储、控制四个模块。本文主要对指纹采集模块、算法模块进行研究，在重点介绍指纹识别算法各个功能的FPGA硬件电路实现方法及效果图，指纹采集模块的SPI接口设计的同时，简单介绍了其它模块的功能以及模块间相互的通信接口设计，并在最后介绍了系统验证平台的搭建以及验证结果。
实验结果表明，基于FPGA硬件方式实现的指纹识别系统能依次顺利完成指纹采集、图像增强、特征点提取以及指纹比对各个功能，完全能满足实际应用的要求。本论文的完成为指纹识别系统的ASIC实现打下了基础。
The fingerprint identification technology has been applied widely as a modern technology in biometrics because of the advantages of uniqueness, universality, irreproducibility. The disadvantages of portability and cubage in traditional Automatic Fingerprint Identification System (AFIS) became more and more obvious while the requirement of safety in mini-equipment was increasing day by day. That provides a huge market in the research of AFIS in the mini-equipments .This dissertation describes the realization of AFIS with Field Programmed Gate Array (FPGA) based on BST FingerAPI1.0 in the background of above. The topics of this dissertation are presented below:
1. After the in-depth research of the characters, applications and tendency of fingerprint identification technology based on DSP or ARM, solutions are brought forward to surmount the disadvantages of those traditional AFIS.
2. A new arithmetic compliant for FPAG in hardware circuit is brought out based on the research of arithmetic existing especially AFIS and BST FingerAPIl.O revision. The new arithmetic is composed of fingerprint gathering, image enhancement, minutiae extraction, minutiae matching, and elides the processes of thinning and binarization compared to the traditional arithmetic. The operational speed is increased while the system performance ability keeps the same.
3. The configuration and function design of AFIS based on FPGA is brought forward, and divide the AFIS to 4 parts as gathering module, arithmetic module, memory module, control module. Gathering module and arithmetic module are particularly introduced in this dissertation, especially the hardware circuits of FPGA, SPI interface of gathering module, the communication interface design between the modules and the function of the other modules. In the end of this dissertation, the validation flat and result of this AFIS are introduced.
The system result of experimentations indicates that the hardware based on FPGA could complete the functions of fingerprint gathering, image enhancement, minutiae extraction, minutiae matching in turn. The completion of this dissertation grounds for the ASIC of fingerprint identification system.
1、在整理学习大量基于DSP或ARM的指纹识别技术资料的基础上，对这类指纹识别技术特点、应用领域和发展趋势做了研究，并针对传统的嵌入式指纹识别系统的缺点提出了相应的解决方案。
2、深入调研了现有各种指纹识别算法，特别是应用在嵌入式AFIS的算法，并结合BST FingerAPI1.0版本算法提出了适用于嵌入FPGA硬件电路的指纹识别算法。该算法主要由指纹检测采集、图像增强、指纹特征点提取以及比对四个部分组成，与传统的算法相比，减少了指纹细化和二值化的过程。这种算法的改进在保证指纹识别系统性能的同时减少了FPGA资源占有量，提高了算法运算速度。
3、对基于FPGA的指纹识别系统进行结构和功能设计，把整个AFIS划分为指纹采集、算法、存储、控制四个模块。本文主要对指纹采集模块、算法模块进行研究，在重点介绍指纹识别算法各个功能的FPGA硬件电路实现方法及效果图，指纹采集模块的SPI接口设计的同时，简单介绍了其它模块的功能以及模块间相互的通信接口设计，并在最后介绍了系统验证平台的搭建以及验证结果。
实验结果表明，基于FPGA硬件方式实现的指纹识别系统能依次顺利完成指纹采集、图像增强、特征点提取以及指纹比对各个功能，完全能满足实际应用的要求。本论文的完成为指纹识别系统的ASIC实现打下了基础。
The fingerprint identification technology has been applied widely as a modern technology in biometrics because of the advantages of uniqueness, universality, irreproducibility. The disadvantages of portability and cubage in traditional Automatic Fingerprint Identification System (AFIS) became more and more obvious while the requirement of safety in mini-equipment was increasing day by day. That provides a huge market in the research of AFIS in the mini-equipments .This dissertation describes the realization of AFIS with Field Programmed Gate Array (FPGA) based on BST FingerAPI1.0 in the background of above. The topics of this dissertation are presented below:
1. After the in-depth research of the characters, applications and tendency of fingerprint identification technology based on DSP or ARM, solutions are brought forward to surmount the disadvantages of those traditional AFIS.
2. A new arithmetic compliant for FPAG in hardware circuit is brought out based on the research of arithmetic existing especially AFIS and BST FingerAPIl.O revision. The new arithmetic is composed of fingerprint gathering, image enhancement, minutiae extraction, minutiae matching, and elides the processes of thinning and binarization compared to the traditional arithmetic. The operational speed is increased while the system performance ability keeps the same.
3. The configuration and function design of AFIS based on FPGA is brought forward, and divide the AFIS to 4 parts as gathering module, arithmetic module, memory module, control module. Gathering module and arithmetic module are particularly introduced in this dissertation, especially the hardware circuits of FPGA, SPI interface of gathering module, the communication interface design between the modules and the function of the other modules. In the end of this dissertation, the validation flat and result of this AFIS are introduced.
The system result of experimentations indicates that the hardware based on FPGA could complete the functions of fingerprint gathering, image enhancement, minutiae extraction, minutiae matching in turn. The completion of this dissertation grounds for the ASIC of fingerprint identification system.
引文
[1] 杨强，谭礼俊．生物识别技术对比浅析．大众科技．2005，2：27-30．
[5] Lin H, Anil J. Integrating faces and fingerprints for personal identification[J]. IEEE Trans.PAMI, 1998, 20(12): 1295-1307
[2] http://www.biometricgroup.com/reports/public/market_report.html.
[3] 李海雄，刘国清．活体指纹识别系统及其应用．计算技术与自动化，2003．
[4] Lin Hong' Ph.D.thesis Automatic personal Identification Using Fingerfrints.
[5] http://www.cse.msu.edu/rgroup/prip.
[6] Jain A, Hong L, Bolle R. On-line fingerprint verification [J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1997, 19 (4): 302-313.
[7] Rao C, Black K. Type classification of fingerprints: a syntactic approach [J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1980, 2 (3): 223-231.
[8] 张成，周媛媛，林嘉宇．指纹采集技术及其产品发展趋势．电子技术应用。2004．
[9] Mehtre B M, Chatter J. Segmentation of fingerprint images a composite method [J]. Pattern Recognition, 1989, 1 (4): 381-385.
[10] Hong L, Wang Y, Jain A. Fingerprint image enhancement: algorithm and performance evaluation [J]. IEEE PAMI, 1998, 20 (8): 777-789.
[11] 罗希平，田捷．自动指纹识别系统中的图像增强和细节匹配算法[J]．软件学报，2002，13(5)：946-956．
[12] 刘勇，晏蒲柳，郭成城．指纹图像脊线提取[J]．武汉大学报，2001，47(1)：106-110．
[13] Maio D, Maltoni D. Direct Gray2Scale Minutiae Detection In Fingerprint [J]. IEEE Transactions on Pattern Analysis and Matching Intelligence, 1997, 19 (1): 27-39.
[14] 张洪光，刘雪梅．指纹识别中的一种局部向量匹配算法[J]．计算机工程，2002，28(4)：106-108
[15] 李志敏，彭志刚．基于动态全局特征的指纹匹配算法的研究[J]．沈阳化工学院学报，2000，14(4)：292-295
[16] 赵娟，王典洪．基于DSP的指纹识别系统原理及应用．安徽电气工程职业技术学院学报，2004，(3)
[17] 陈庆接，李见为，张腾，王玮．基于ARM9的嵌入式自动指纹识别系统．重庆大学学报．2004(9)．
[18] Dinesh P M, Eam K T. An automated matching technique for fingerprint identification [J]. First International Conference on Knowledge-Based Intelligent Electronic Systems, 1997, 5(1): 21-23
[19] 王崇文，李见为，周宏文等．指纹识别系统的设计与实现．计算机应用，2001．
[20] Xin jian, Chen Jie, Tian Jiangang, Cheng Xin. Segmentation of fingerprint images using linear classifier [J]. EURASIP Journal on Applied Signal Processing, 2004, 2004(4): 480-494.
[21] 唐良瑞，谢晓辉，蔡安妮，孙景鳌．基于D-S证据理论的指纹图像分割方法[J]．计算机学报，2003，26(7)：887-892．
[22] Verma M R, Majumdar A K, Chatterjee B. Edge detection in fingerprints [J]. Pattern Recongition, 1987, (5): 513-523.
[23] 耿茵茵，唐良瑞，指纹图像分级分割算法，北方工业大学学报，2000，12(3)：17-20．
[24] 杨海军，梁德群．基于多尺度屋顶边缘检测的指纹图像分割方法．中国图像图形学报，2005，1：42-46．
[25] Sherlock D, Momro D M, Millard K. Fingerprint enhancement by directional Fourier filter [J]. IEEE Proceedings of Vision Image and Signal Processing.
[26] Lin Hong, Yi fei Wan, Anil Jain. Fingerprint Image Enhancement: Algorithm and Performance Evaluation. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1998 Vol. 20 (1): 777-788.
[27] 尹义龙，詹小四，谭台哲等．基于加博函数的指纹增强算法及其应用[J]．软件学报，2003，14(3)：484-489．
[28] B.G.Sherlock, D.M.Monro, Fingerprint Enhancement by Directional Fourier Filtering, IEEE Proc. Vis. Image Signal Processing, 1994, 141 (2): 87-94.
[29] O'Gorman L, Nickersion J V. An approach to fingerprint filter design [J]. Pattern Recognition, 1989, 22 (1): 29-38.
[30] 简兵，庄镇泉，李军．一种新颖的指纹细节提取算法[A]．第6届国际青年计算机工作者会议暨第二届生物特征识别研讨会[C]，2001，S4．7-13．
[31] Jiang X, Yau WY, Ser W. Detecting the fingerprint minutiae by adaptive tracing the gray level ridge [J]. Pattern Recognition, 2001, 34: 999-1013.
[32] 姚文庆，肖雪静．指纹特征匹配方法的研究与实现[J]．天津大学学报，1992，4(4)：48-55．
[33] Chen Z, Kuo C H. A toplogy based matching-algorithm fingerprint authentication [J]. IEEE International Carhanan Conference on Security Technology, 1991, 31 (2): 84-87.
[34] Virtex-4 FPGA Family: Complete Data Sheet [Z]. 2006
[35] 王波涛，孙景鳌，蔡安妮等．指纹取像技术．北方工业大学学报，2001，7．
[36] 王崇文，赵德申，王延才等．指纹取像与指纹识别．计算机工程与应用，2002，3．
[37] MBF200 Solid State Fingerprint Sensor, Data Sheet[Z]. 2002
[5] Lin H, Anil J. Integrating faces and fingerprints for personal identification[J]. IEEE Trans.PAMI, 1998, 20(12): 1295-1307
[2] http://www.biometricgroup.com/reports/public/market_report.html.
[3] 李海雄，刘国清．活体指纹识别系统及其应用．计算技术与自动化，2003．
[4] Lin Hong' Ph.D.thesis Automatic personal Identification Using Fingerfrints.
[5] http://www.cse.msu.edu/rgroup/prip.
[6] Jain A, Hong L, Bolle R. On-line fingerprint verification [J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1997, 19 (4): 302-313.
[7] Rao C, Black K. Type classification of fingerprints: a syntactic approach [J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1980, 2 (3): 223-231.
[8] 张成，周媛媛，林嘉宇．指纹采集技术及其产品发展趋势．电子技术应用。2004．
[9] Mehtre B M, Chatter J. Segmentation of fingerprint images a composite method [J]. Pattern Recognition, 1989, 1 (4): 381-385.
[10] Hong L, Wang Y, Jain A. Fingerprint image enhancement: algorithm and performance evaluation [J]. IEEE PAMI, 1998, 20 (8): 777-789.
[11] 罗希平，田捷．自动指纹识别系统中的图像增强和细节匹配算法[J]．软件学报，2002，13(5)：946-956．
[12] 刘勇，晏蒲柳，郭成城．指纹图像脊线提取[J]．武汉大学报，2001，47(1)：106-110．
[13] Maio D, Maltoni D. Direct Gray2Scale Minutiae Detection In Fingerprint [J]. IEEE Transactions on Pattern Analysis and Matching Intelligence, 1997, 19 (1): 27-39.
[14] 张洪光，刘雪梅．指纹识别中的一种局部向量匹配算法[J]．计算机工程，2002，28(4)：106-108
[15] 李志敏，彭志刚．基于动态全局特征的指纹匹配算法的研究[J]．沈阳化工学院学报，2000，14(4)：292-295
[16] 赵娟，王典洪．基于DSP的指纹识别系统原理及应用．安徽电气工程职业技术学院学报，2004，(3)
[17] 陈庆接，李见为，张腾，王玮．基于ARM9的嵌入式自动指纹识别系统．重庆大学学报．2004(9)．
[18] Dinesh P M, Eam K T. An automated matching technique for fingerprint identification [J]. First International Conference on Knowledge-Based Intelligent Electronic Systems, 1997, 5(1): 21-23
[19] 王崇文，李见为，周宏文等．指纹识别系统的设计与实现．计算机应用，2001．
[20] Xin jian, Chen Jie, Tian Jiangang, Cheng Xin. Segmentation of fingerprint images using linear classifier [J]. EURASIP Journal on Applied Signal Processing, 2004, 2004(4): 480-494.
[21] 唐良瑞，谢晓辉，蔡安妮，孙景鳌．基于D-S证据理论的指纹图像分割方法[J]．计算机学报，2003，26(7)：887-892．
[22] Verma M R, Majumdar A K, Chatterjee B. Edge detection in fingerprints [J]. Pattern Recongition, 1987, (5): 513-523.
[23] 耿茵茵，唐良瑞，指纹图像分级分割算法，北方工业大学学报，2000，12(3)：17-20．
[24] 杨海军，梁德群．基于多尺度屋顶边缘检测的指纹图像分割方法．中国图像图形学报，2005，1：42-46．
[25] Sherlock D, Momro D M, Millard K. Fingerprint enhancement by directional Fourier filter [J]. IEEE Proceedings of Vision Image and Signal Processing.
[26] Lin Hong, Yi fei Wan, Anil Jain. Fingerprint Image Enhancement: Algorithm and Performance Evaluation. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1998 Vol. 20 (1): 777-788.
[27] 尹义龙，詹小四，谭台哲等．基于加博函数的指纹增强算法及其应用[J]．软件学报，2003，14(3)：484-489．
[28] B.G.Sherlock, D.M.Monro, Fingerprint Enhancement by Directional Fourier Filtering, IEEE Proc. Vis. Image Signal Processing, 1994, 141 (2): 87-94.
[29] O'Gorman L, Nickersion J V. An approach to fingerprint filter design [J]. Pattern Recognition, 1989, 22 (1): 29-38.
[30] 简兵，庄镇泉，李军．一种新颖的指纹细节提取算法[A]．第6届国际青年计算机工作者会议暨第二届生物特征识别研讨会[C]，2001，S4．7-13．
[31] Jiang X, Yau WY, Ser W. Detecting the fingerprint minutiae by adaptive tracing the gray level ridge [J]. Pattern Recognition, 2001, 34: 999-1013.
[32] 姚文庆，肖雪静．指纹特征匹配方法的研究与实现[J]．天津大学学报，1992，4(4)：48-55．
[33] Chen Z, Kuo C H. A toplogy based matching-algorithm fingerprint authentication [J]. IEEE International Carhanan Conference on Security Technology, 1991, 31 (2): 84-87.
[34] Virtex-4 FPGA Family: Complete Data Sheet [Z]. 2006
[35] 王波涛，孙景鳌，蔡安妮等．指纹取像技术．北方工业大学学报，2001，7．
[36] 王崇文，赵德申，王延才等．指纹取像与指纹识别．计算机工程与应用，2002，3．
[37] MBF200 Solid State Fingerprint Sensor, Data Sheet[Z]. 2002
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |