基于局部生长融合的指纹匹配算法研究
|
摘要
指纹是存在于人手指端部的纹理,它具有终身的不变性和唯一性的特点,在身份认证方面有着一百多年的历史。指纹首先在公安刑侦中得到广泛的应用。随着计算机的不断发展和普及,社会安全对于身份认证的要求不断提高,自动指纹匹配技术吸引了众多企业和研究机构投入了大量的研究力量。如今自动指纹匹配技术不但广泛应用于公安刑侦,而且在民众日常生活中应用也日益普遍。自动指纹匹配
近年来,随着科技发展水平的不断提高,身份验证已经成为生产生活中不可或缺的重要环节。其中的生物特征识别技术更是以其出众的安全性、可靠性、便携性,成为该技术领域的研究热点,同时也吸引了众多知名企业投入研发力量致力于生物特征识别技术的产业化。
指纹识别技术,作为生物特征识别中历史最悠久、应用最为广泛的主流组成部分,已经得到了长足的发展;但是在指纹识别领域仍然存在很多极具挑战性的问题,例如指纹图像的非线性形变、指纹不可靠细节特征匹配、低质量图像增强等。在本文中,重点对仅利用指纹细节点的指纹匹配方法进行了深入研究,提出了一种新颖算法,并在公开数据库上进行了测试。本文的主要贡献在于:
首先,提出了一种基于指纹应用环境的局部结构相容性定义。该相容性定义巧妙地将指纹细节点局部结构联系起来,较为准确地描述了局部结构之间的相对空间位置关系在经历某次变换前后的保持一致性的程度,为此后的局部结构相似度调节提供了依据。
其次对于单纯利用指纹细节点进行匹配的情况,提出了一种基于指纹细节点局部结构相容性的指纹匹配方法。该算法仅利用细节点特征,通过构造细节点局部结构来建立局部结构相似度矩阵;并采用了以局部结构相容性思想为基础的松弛过程,对局部结构相似度矩阵进行调节,使其中真实匹配结构对的相似度逐渐突出。最后利用扩展搜索的方法对匹配结果进行扩展。此算法克服了以往基于细节的匹配方法单纯依靠某一组线性变换校准参数的缺点,在一定程度上能够容忍非线性变换的存在,有效地提高了算法在质量较差的图像上的性能。在FVC2004数据库上的试验表明,该方法与其他同类单纯使用细节点特征的算法相比,有效地提高整体性能。
Recently, personal identification has been one of the most indispensable parts in people’s daily life with the development of the science and technology. As one of the important parts, Biometrics has drawn more and more research interest and industrial focus due to its high safety, stability, and convenience.
As the most famous and widely used technology, though fingerprint identification has received more attention and been studied for a long time, there are still some unsolved challenging problems: such as non-linear distortion in fingerprint images, fingerprint matching based on unstable minutiae, low-quality image enhancement and so on. In this paper, we set focus on fingerprint matching only using minutiae, and propose a novel fingerprint matching algorithm based on local structure compatibility, which has been evaluated on FVC database. The main contribution to fingerprint matching in this work includes:
First, local structure compatibility was proposed, which describes the mutual relationship between the neighboring local structures. And this compatibility could be thought as the stability of the neighboring structures’relative positions during a certain transformation. The compatibility is also the basis of the local structure’s similarity adjustment.
Secondly, a novel fingerprint matching algorithm based on local structure compatibility was proposed. The algorithm only uses the minutiae to construct local structures and form the similarity matrix (TSM) between two images. Based on the compatibility, we adjust the similarity of each matching pair according to their neighbors’matching status in order to make the genuine matching pair distinct from the others. At last, we use an extended searching step to gain more pairs so as to avoid simply depending on only one reference pair or transformation parameter. The evaluation on the FVC2004 database demonstrated improvement compared to the congeneric methods.
近年来,随着科技发展水平的不断提高,身份验证已经成为生产生活中不可或缺的重要环节。其中的生物特征识别技术更是以其出众的安全性、可靠性、便携性,成为该技术领域的研究热点,同时也吸引了众多知名企业投入研发力量致力于生物特征识别技术的产业化。
指纹识别技术,作为生物特征识别中历史最悠久、应用最为广泛的主流组成部分,已经得到了长足的发展;但是在指纹识别领域仍然存在很多极具挑战性的问题,例如指纹图像的非线性形变、指纹不可靠细节特征匹配、低质量图像增强等。在本文中,重点对仅利用指纹细节点的指纹匹配方法进行了深入研究,提出了一种新颖算法,并在公开数据库上进行了测试。本文的主要贡献在于:
首先,提出了一种基于指纹应用环境的局部结构相容性定义。该相容性定义巧妙地将指纹细节点局部结构联系起来,较为准确地描述了局部结构之间的相对空间位置关系在经历某次变换前后的保持一致性的程度,为此后的局部结构相似度调节提供了依据。
其次对于单纯利用指纹细节点进行匹配的情况,提出了一种基于指纹细节点局部结构相容性的指纹匹配方法。该算法仅利用细节点特征,通过构造细节点局部结构来建立局部结构相似度矩阵;并采用了以局部结构相容性思想为基础的松弛过程,对局部结构相似度矩阵进行调节,使其中真实匹配结构对的相似度逐渐突出。最后利用扩展搜索的方法对匹配结果进行扩展。此算法克服了以往基于细节的匹配方法单纯依靠某一组线性变换校准参数的缺点,在一定程度上能够容忍非线性变换的存在,有效地提高了算法在质量较差的图像上的性能。在FVC2004数据库上的试验表明,该方法与其他同类单纯使用细节点特征的算法相比,有效地提高整体性能。
Recently, personal identification has been one of the most indispensable parts in people’s daily life with the development of the science and technology. As one of the important parts, Biometrics has drawn more and more research interest and industrial focus due to its high safety, stability, and convenience.
As the most famous and widely used technology, though fingerprint identification has received more attention and been studied for a long time, there are still some unsolved challenging problems: such as non-linear distortion in fingerprint images, fingerprint matching based on unstable minutiae, low-quality image enhancement and so on. In this paper, we set focus on fingerprint matching only using minutiae, and propose a novel fingerprint matching algorithm based on local structure compatibility, which has been evaluated on FVC database. The main contribution to fingerprint matching in this work includes:
First, local structure compatibility was proposed, which describes the mutual relationship between the neighboring local structures. And this compatibility could be thought as the stability of the neighboring structures’relative positions during a certain transformation. The compatibility is also the basis of the local structure’s similarity adjustment.
Secondly, a novel fingerprint matching algorithm based on local structure compatibility was proposed. The algorithm only uses the minutiae to construct local structures and form the similarity matrix (TSM) between two images. Based on the compatibility, we adjust the similarity of each matching pair according to their neighbors’matching status in order to make the genuine matching pair distinct from the others. At last, we use an extended searching step to gain more pairs so as to avoid simply depending on only one reference pair or transformation parameter. The evaluation on the FVC2004 database demonstrated improvement compared to the congeneric methods.
引文
[1] D. Maltoni, D. Maio, A.K. Jain, and S. Prabhakar, Handbook of Fingerprint Recognition. New York,Springer, 2003.
[2] R. Cappelli, Alessandra Lumini, D. Maio, D. Maltoni. "Fingerprint Classification by Directional Image Partitioning," IEEE Transaction on Pattern Analysis and Machine Intelligence.,vol21-5, pp.402-421 1999
[3] Fingerprint Verification Competition (FVC2004), http://bias.csr.unibo.it/fvc2004, 2004
[4] Fingerprint Verification Competition (FVC2002), http://bias.csr.unibo.it/fvc2002/, 2002
[5] Fingerprint Verification Competition (FVC2000), http://bias.csr.unibo.it/fvc2000/, 2000
[6] NIST fingerprint database, http://www.nist.gov/data/
[7] Sharath Pankanti, Salil Prabhakar, and Anil K. Jain, “On the Individuality of Fingerprints”, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol24-8, pp.1010-1025, 2002
[8] Crouzil,Massip-Pailhes, Castan, “A New Correlation Criterion Based on Gradient Fields Similarity” Proceeding of International Conference on Pattern Recognition (13th), pp.632-636, 1996
[9] Hatano T, Adachi T, Shigematsu S, Morimura H et al, “A Fingerprint Verification Algorithm using the Differential Matching Rate”, Proceeding of International Conference on Pattern Recognition (16th), Vol3, pp.799-802, 2002
[10] A. Bazen, Gerben Berwaaijen, Gerez et al, “A Correlation-Based Fingerprint Verification System”, Proceeding of Workshop on Circuits Systems and Signal Processing (ProRISC), pp. 205-213, 2000
[11] Zsolt Kovacs-Vajna, “A Fingerprint Verification System Based on Triangular Matching and Dynamic Time Warping”, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol22-11, pp.1266-1276, 2000
[12] 冈萨雷斯,伍兹,《数字图像处理(第二版)》,电子工业出版社,2004
[13] Chang S.H, Cheng F.H, Hsu W.H, “Fast Algorithm for Point Pattern-Matching: Invariant to Transformations, Rotations and Scale Changes”, Pattern Recognition,Vol.30-2, pp.311-320, 1997
[14] S. Ranade, A. Rosenfeld, "Point Pattern Matching by Relaxation," Pattern Recognition, vol.12, pp.269-275, 1980
[15] A. Rosenfeld, R. Hummel, and S. Zucker, "Scene Labeling by Relaxation Operations," IEEE Trans. Sys, Man and Cyber, vol.SMC-6, no.6, 1976
[16] Jain, A, and Lin Hong, "on-line fingerprint verification," IEEE Transaction on Pattern Analysis and Machine Intelligence., vol19, no4, pp.302-314 1997
[17] Xifeng Tong, Jianhua Huang, Xianglong Tang, Daming Shi, “Fingerprint Minutiae Matching using the Adjacent Feature Vector”, Pattern Recognition Letters, Vol26, pp.1337-1345, 2005
[18] En Zhu, Jianping Yin, Guomin Zhang, “Fingerprint Matching Based on global Alignment of Multiple Reference Minutiae”, Pattern Recognition, Vol 38, pp.1685-1694, 2005
[19] Jin Qi, Suzhen Yang, Yangsheng Wang, “Fingerprint Matching Combining the Global Orientation Field with Minutia”, Pattern Recognition Letters, Vol26, pp.2424-2430, 2005
[20] Xudong. Jiang, Wei-Yun Yau, "Fingerprint minutiae matching based on the local and global structures," ICPR2000, Vol.2, pp: 1038-1041, 2000
[21] Ratha. N.K, Bolle. R.M, Pandit. V.D and Vaish. V, "Robust fingerprint authentication using local structural similarity," 15th IEEE Workshop on Applications of Computer Vision, 29-34 2000
[22] Bebis. G, Deaconu. T, and Georgiopoulos. M, "Fingerprint identification using Delaunay triangulation," International Conference on Information Intelligence and Systems, 452-459, 1999
[23] Giuseppe Parziale, Albert Niel, “A Fingerprint Matching Using Minutiae Triangulation”, ICBA 2004, LNCS 3072, pp. 241–248, 2004.
[24] Huimin Deng, Qiang Huo, "Minutiae Matching Based Fingerprint Verification Using Delaunay Triangulation and Aligned-Edge-Guided Triangle Matching," AVBPA2005, vol 3546/2005, pp270, 2005
[25] G. Parziale, A. Niel, "A Fingerprint Matching Using Minutiae Triangulation," ICBA2004, vol 3072/2004, pp241-248, 2004
[26] Jin Qi, Yangsheng Wang, "A Robust Fingerprint Matching Method", Pattern Recognition, vol 38 (2005), pp1665-1671, 2005
[27] Xinjian Chen, Jie Tian, and Xin Yang, "A Novel Algorithm for Distorted Fingerprint Matching Based on Fuzzy Features Match," AVBPA2005, vol 3546/2005, pp665-673, 2005
[28] R. Cappelli, D. Maio, D. Maltoni, J.Wayman, A.K. Jain."Performance Evaluation of Fingerprint Verification Systems," IEEE Trans. PAMI. vol28-1, pp3-18 2006
[29] Yuliang He, Jie Tian, Liang Li, Hong Chen, Xin Yang, "Fingerprint Matching Based on Global Comprehensive Similarity," IEEE Trans. PAMI. Vol29-6, pp850-861 2006
[30] Fang-Hsuan Cheng, "Point Pattern Matching Algorithm Invariant to Geometrical Transformation and Distortion", Pattern Recognition Letters, Vol17, pp.1429-1435, 1996
[31] 张堂辉,指纹匹配算法的研究与应用,中国科学院自动化研究所,博士学位论文,2003
[32] 漆进,指纹匹配算法研究及应用,中国科学院自动化研究所,博士学位论文,2005
[2] R. Cappelli, Alessandra Lumini, D. Maio, D. Maltoni. "Fingerprint Classification by Directional Image Partitioning," IEEE Transaction on Pattern Analysis and Machine Intelligence.,vol21-5, pp.402-421 1999
[3] Fingerprint Verification Competition (FVC2004), http://bias.csr.unibo.it/fvc2004, 2004
[4] Fingerprint Verification Competition (FVC2002), http://bias.csr.unibo.it/fvc2002/, 2002
[5] Fingerprint Verification Competition (FVC2000), http://bias.csr.unibo.it/fvc2000/, 2000
[6] NIST fingerprint database, http://www.nist.gov/data/
[7] Sharath Pankanti, Salil Prabhakar, and Anil K. Jain, “On the Individuality of Fingerprints”, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol24-8, pp.1010-1025, 2002
[8] Crouzil,Massip-Pailhes, Castan, “A New Correlation Criterion Based on Gradient Fields Similarity” Proceeding of International Conference on Pattern Recognition (13th), pp.632-636, 1996
[9] Hatano T, Adachi T, Shigematsu S, Morimura H et al, “A Fingerprint Verification Algorithm using the Differential Matching Rate”, Proceeding of International Conference on Pattern Recognition (16th), Vol3, pp.799-802, 2002
[10] A. Bazen, Gerben Berwaaijen, Gerez et al, “A Correlation-Based Fingerprint Verification System”, Proceeding of Workshop on Circuits Systems and Signal Processing (ProRISC), pp. 205-213, 2000
[11] Zsolt Kovacs-Vajna, “A Fingerprint Verification System Based on Triangular Matching and Dynamic Time Warping”, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol22-11, pp.1266-1276, 2000
[12] 冈萨雷斯,伍兹,《数字图像处理(第二版)》,电子工业出版社,2004
[13] Chang S.H, Cheng F.H, Hsu W.H, “Fast Algorithm for Point Pattern-Matching: Invariant to Transformations, Rotations and Scale Changes”, Pattern Recognition,Vol.30-2, pp.311-320, 1997
[14] S. Ranade, A. Rosenfeld, "Point Pattern Matching by Relaxation," Pattern Recognition, vol.12, pp.269-275, 1980
[15] A. Rosenfeld, R. Hummel, and S. Zucker, "Scene Labeling by Relaxation Operations," IEEE Trans. Sys, Man and Cyber, vol.SMC-6, no.6, 1976
[16] Jain, A, and Lin Hong, "on-line fingerprint verification," IEEE Transaction on Pattern Analysis and Machine Intelligence., vol19, no4, pp.302-314 1997
[17] Xifeng Tong, Jianhua Huang, Xianglong Tang, Daming Shi, “Fingerprint Minutiae Matching using the Adjacent Feature Vector”, Pattern Recognition Letters, Vol26, pp.1337-1345, 2005
[18] En Zhu, Jianping Yin, Guomin Zhang, “Fingerprint Matching Based on global Alignment of Multiple Reference Minutiae”, Pattern Recognition, Vol 38, pp.1685-1694, 2005
[19] Jin Qi, Suzhen Yang, Yangsheng Wang, “Fingerprint Matching Combining the Global Orientation Field with Minutia”, Pattern Recognition Letters, Vol26, pp.2424-2430, 2005
[20] Xudong. Jiang, Wei-Yun Yau, "Fingerprint minutiae matching based on the local and global structures," ICPR2000, Vol.2, pp: 1038-1041, 2000
[21] Ratha. N.K, Bolle. R.M, Pandit. V.D and Vaish. V, "Robust fingerprint authentication using local structural similarity," 15th IEEE Workshop on Applications of Computer Vision, 29-34 2000
[22] Bebis. G, Deaconu. T, and Georgiopoulos. M, "Fingerprint identification using Delaunay triangulation," International Conference on Information Intelligence and Systems, 452-459, 1999
[23] Giuseppe Parziale, Albert Niel, “A Fingerprint Matching Using Minutiae Triangulation”, ICBA 2004, LNCS 3072, pp. 241–248, 2004.
[24] Huimin Deng, Qiang Huo, "Minutiae Matching Based Fingerprint Verification Using Delaunay Triangulation and Aligned-Edge-Guided Triangle Matching," AVBPA2005, vol 3546/2005, pp270, 2005
[25] G. Parziale, A. Niel, "A Fingerprint Matching Using Minutiae Triangulation," ICBA2004, vol 3072/2004, pp241-248, 2004
[26] Jin Qi, Yangsheng Wang, "A Robust Fingerprint Matching Method", Pattern Recognition, vol 38 (2005), pp1665-1671, 2005
[27] Xinjian Chen, Jie Tian, and Xin Yang, "A Novel Algorithm for Distorted Fingerprint Matching Based on Fuzzy Features Match," AVBPA2005, vol 3546/2005, pp665-673, 2005
[28] R. Cappelli, D. Maio, D. Maltoni, J.Wayman, A.K. Jain."Performance Evaluation of Fingerprint Verification Systems," IEEE Trans. PAMI. vol28-1, pp3-18 2006
[29] Yuliang He, Jie Tian, Liang Li, Hong Chen, Xin Yang, "Fingerprint Matching Based on Global Comprehensive Similarity," IEEE Trans. PAMI. Vol29-6, pp850-861 2006
[30] Fang-Hsuan Cheng, "Point Pattern Matching Algorithm Invariant to Geometrical Transformation and Distortion", Pattern Recognition Letters, Vol17, pp.1429-1435, 1996
[31] 张堂辉,指纹匹配算法的研究与应用,中国科学院自动化研究所,博士学位论文,2003
[32] 漆进,指纹匹配算法研究及应用,中国科学院自动化研究所,博士学位论文,2005
© 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 |