自动指纹识别系统（automatic fingerprint identification system，AFIS）已经广泛应用于签证、海关、刑侦等领域.为逃避基于AFIS的身份识别，有些犯罪分子通过切割、移植、灼烧等手段改变其指纹的脊结构，由此得到的指纹被称为形变指纹（altered fingerprint）.目前AFIS并不能有效地检测形变指纹.于是提出一种基于方向场不连续性和方向场差异图不连续性的形变指纹检测方法，并提出了用于提取这种不连续性的算子.首先计算指纹的方向场及其差异图，并对差异图进行增强.然后用所提出的算子分别从方向场和增强后的差异图中提取不连续特征.最后基于不连续特征，用支持向量机对形变指纹或真实指纹进行分类.此外，还构建了形变指纹数据库，并在该库上测试所提方法的性能.结果表明，所提方法能够有效识别形变指纹.
Automatic fingerprint identification system (AFIS) has been widely used in visas, customs, criminal investigations. In order to avoid AFIS-based identification, some criminals change the ridge structure of their fingerprints by cutting, transplanting, burning, etc. The resulting fingerprints are called altered fingerprint. At present, AFIS can't effectively detect altered fingerprints. In this paper, an altered fingerprint detection method based on orientation field discontinuity and orientation field difference map discontinuity is proposed, and an operator for extracting this discontinuity is proposed. Firstly, the orientation field of fingerprint and its difference map are calculated, and the difference map is enhanced. Then, the proposed operator extracts the discontinuous features from the orientation field and the enhanced difference map. Finally, based on the discontinuous features, altered fingerprints and real fingerprints are classified using support vector machines (SVM). In addition, this paper also builds an altered fingerprint database, and tests the performance of the method on this library. The results show that the proposed scheme can detect the altered fingerprint effectively.
[1] Cummins H. Attempts to alter and obliterate finger-prints[J]. Journal of Criminal Law and Criminology, 1935, 25(6):982-991.
[2] Wertheim K. An extreme case of fingerprint mutilation[J]. Journal of Forensic Identification, 1998, 48(4):466-477.
[3] Feng J, Jain A K, Ross A. Detecting altered fingerprints[C]//20th International Conference on Pattern Recognition (ICPR). IEEE, 2010:1622-1625.
[4] Petrovici A, Lazar C. Identifying fingerprint alteration using the reliability map of the orientation field[J]. The Annals of the Univeristy of Craiova. Series:Automation, Computers, Electronics and Mechatronics, 2010, 7(34):45-52.
[5] Yoon S, Feng J, Jain A K. Altered fingerprints:analysis and detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(3):451-464.
[6] Yoon S, Jain A K. Is there a fingerprint pattern in the image?[C]//International Conference on Biometrics (ICB). IEEE, 2013:1-8.
[7] Tiribuzi M, Pastorelli M, Valigi P. A multiple kernel learning framework for detecting altered fingerprints[C]//Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012). IEEE, Tsukuba, Japan, 2012:3402-3405.
[8] Ellingsgaard J, Sousedik C, Busch C. Detecting fingerprint alterations by orientation field and minutiae orientation analysis[C]//2nd International Workshop on Biometrics and Forensics. IEEE, Valletta, Malta, 2014:1-6.
[9] Ellingsgaard J, Busch C. Altered fingerprint detection[M]//Handbook of Biometrics for Forensic Science. Springer, 2017:85-123.
[10] Haraksim R, Anthonioz C, Champod C. Altered fingerprint detection-algorithm performance evaluation[C]//4th International Conference on Biometrics and Forensics (IWBF). IEEE, Limassol, Cyprus, 2016:1-6.
[11] Lazar C, Psenita C D, Tiganasu A. Comparative analysis of orientation field descriptors used for detecting altered fingerprints[C]//20th International Conference on System Theory, Control and Computing (ICSTCC). IEEE, Sinaia, Romania, 2016:715-720.
[12] Hong L, Wan Y, Jain A. Fingerprint image enhancement:algorithm and performance evaluation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20(8):777-789.
[13] Gu J, Zhou J, Zhang D. A combination model for orientation field of fingerprints[J]. Pattern Recognition, 2004, 37(3):543-553
[14] Watson C I, Wilson C L. Nist special database 4[DB]. Fingerprint Database, National Institute of Standards and Technology.
[15] Chang C C, Lin C J. LIBSVM:a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3):27.
