摘要
近年来，以指纹为代表的生物特征识别技术引起了人们的广泛关注。指纹识别技术以其可靠性、稳定性在生物特征识别领域得到了广泛的应用。准确、可靠地特征提取是确保自动指纹识别系统性能的前提和基础，而指纹增强和匹配对指纹特征提取的准确性和指纹最终的匹配结果有着直接的影响。本文结合加博滤波，对指纹增强和纹理匹配方法进行了研究。
本文的主要工作分为指纹增强和纹理匹配两个方面，其中指纹增强包括综合使用指纹质量和加博滤波的指纹增强，综合使用指纹指纹码特征向量和加博滤波的指纹增强。纹理匹配包括传统纹理匹配的实现，纹理匹配算法的改进和使用指纹质量的纹理匹配。
综合使用指纹质量和加博滤波的指纹增强：本方法采用自适应方法估计指纹图像的平均纹线距离作为加博滤波器的参数，并将图像质量测量评价结果应用于指纹增强，提出了一种综合使用图像质量和加博滤波的指纹增强方法。首先，利用大窗口频谱分析方法估计指纹的平均纹线频率，构造八个不同方向的加博滤波器模板；然后，利用加博滤波器分别对原始指纹图像进行滤波处理，得到八幅滤波后图像；最后，利用指纹图像质量评价因子对这八幅图像择优合成最终的增强图像。在典型指纹图像上的实验结果表明，本方法在一定程度上克服了传统的图像分块加博增强方法在指纹模式区性能表现不佳的问题。同时，由于该方法不依赖图像纹线方向，自适应获取纹线频率这一加博滤波参数，对低质量指纹图像表现出较强的鲁棒性。
综合使用指纹码特征向量和加博滤波的指纹增强：采用自适应方法估计指纹图像的平均纹线距离作为加博滤波器的参数，并将指纹码特征向量应用于指纹增强，提出了一种综合使用图像码特征向量和加博滤波的指纹增强方法。首先，利用大窗口频谱分析方法估计指纹的平均纹线频率，构造八个不同方向的加博滤波器模板；然后，利用加博滤波器分别对原始指纹图像进行滤波处理，得到八幅滤波后图像；最后，根据指纹码特征向量获得的数据对这八幅图像择优合成最终的增强图像。实验结果表明，本方法不但具有综合使用指纹质量和加博滤波的指纹增强方法所具有的优点，而且比指纹质量因子更能准确表示指纹纹理清晰程度，有效地减少了增强图像合成中坏块的产生。
纹理匹配：本文实现并改进的基于加博滤波的纹理匹配同时利用指纹的局部特征和指纹全局特征，不单纯依赖局部细节特征，适合细节特征缺少、伪点较多的指纹的匹配，弥补了基于细节点的匹配方法的不足。纹理匹配依靠输入指纹指纹码与模板指纹指纹码之间欧拉的比较，确定两幅指纹是否同源指纹。其中，指纹码的获取：首先，以核心点为中心将这八幅图像划分成扇形区域；然后，构造八个不同方向的加博滤波器模板，利用加博滤波器分别对原始指纹图像进行滤波处理，得到八幅滤波后图像；计算每个扇形区域的指纹码；最后，将八幅图像的扇形区域按照一定的顺序编成指纹码。另外，本文对上述纹理匹配方法进行了一些改进，对指纹核心点比较接近指纹图像边缘，指纹码部分区域在指纹图像有效区域外的情况进行特殊处理，对采偏、有效面积较少的指纹也能正常进行编码。实验表明，本文实现并改进的纹理匹配算法可以有效识别输入指纹。
使用指纹质量特征向量的纹理匹配：将新的特征向量应用于纹理匹配，进行了实验。依靠输入指纹得指纹码与模板指纹指纹码的比较，确定两幅指纹是否同源指纹。在获取指纹质量编码的时候：首先，以核心点为中心将这八幅图像划分成扇形区域；然后，构造八个不同方向的加博滤波器模板，利用加博滤波器分别对原始指纹图像进行滤波处理，得到八幅滤波后图像；计算每个扇形区域的指纹质量特征向量；最后，将八幅图像的扇形区域按照一定的顺序编成指纹码。
Automatic fingerprint identification technique has become a research focus in the fields of biometrics in last decades. And because of its credibility and stability, automatic fingerpnnt identification has been widely used in biometrics. Extract features exactly and credibility is the premise and basics of the automatic fingerprint identification system, moreover fingerpnnt enhancement and matching have direct influence on the veracity of feature extraction and the final matching result. This paper studies fingerprint enhancement and texture matching combining with the gabor filter.
This paper mainly deals with fingerprint enhancement and texture matching. Fingerprint enhancement include: fingerprint enhancement combining fingerprint image quality and gabor filter, fingerprint enhancement combining finger code feature vector and gabor filter. Texture matching include: Implement and improvement the traditional filter based matching, texture matching using fingerprint image quality.
Fingerprint enhancement combining with fingerprint image quality and gabor filter: Estimate the ridge distance of fingerprint based on spectrum analysis with big window and by use fingerprint image quality into enhancement, we proposed fingerpnnt enhancement combining with fingerprint image quality and gabor filter. First, estimate the ridge distance of fingerpnnt based on spectrum analysis with big window, construct eight gabor filters in eight different directions; then filter the image in eight directions and divide the eight filtered images into square sectors ; finally using fingerprint image quality to compose the finally enhanced image. Expenmental results demonstrate that, the method conquers the shortage of traditional enhancement in pattern areas, and is robust to fingerpnnt images of low quality.
Fingerpnnt enhancement: Estimate the ndge distance of fingerpnnt based on spectrum analysis with big window and by use finger code feature vector into enhancement, we proposed fingerprint enhancement combining with finger code feature vector and gabor filter. First, estimate the ndge distance of fingerprint based on spectrum analysis with big window, construct eight gabor filters in eight different directions; then filter the image in eight directions and divide the eight filtered images into square sectors ; finally using finger code feature vector to compose the finally enhanced image. Expenmental results demonstrate that, the method not only possess the advantages of fingerpnnt enhancement combining fingerpnnt image quality and gabor filter, but also reduce the chance of bad blocks in fingerprint images.
Texture matching: For a considerable fraction of population, the representations based on explicit detection of complete ridge structures in the fingerprint are difficult to extract automatically The widely used minutiae-based representation does not utilize a significant component of the rich discriminatory information available in the fingerprints. Local ridge structures cannot be completely characterized by minutiae. The proposed filter-based texture matching uses a bank of Gabor filters to capture both local and global details in a fingerprint as a compact fixed length Finger Code. The fingerprint matching is based on the Euclidean distance between the two corresponding Finger Codes. First determine a reference point and region of interest for the fingerprint image; tessellate the region of interest around the reference point; then filter the region of interest in eight different directions using a bank of Gabor filters; finally compute the average absolute deviation from the mean of gray values in individual sectors in filtered images to define the feature vector or the Finger Code. Expenmental results demonstrate that the improved texture matching can identify the input fingerprints effectively
Texture matching using fingerprint image quality: we try to introduce new feature vector into texture matching. The fingerprint matching is based on the Euclidean distance between the two corresponding Finger Codes. First determine a reference point and region of interest for the fingerprint image; tessellate the region of interest around the reference point; then filter the region of interest in eight different directions using a bank of Gabor filters; finally compute the fingerprint image quality
in individual sectors in filtered images to define the feature vector or the Finger Code.
引文
[1] 尹义龙，宁新宝，张晓梅．自动指纹识别技术的发展与应用[J]．南京大学学报，2002，38(1)：29-35．
[2] J. G. Daugman. High confidence visual recognition of persons by a test of statistical independence [J]. IEEE Transactions on Pattern Recognition and Machine Analysis, 1993, 15(11): 1148-1161.
[3] Jr. J. Campbell. Speaker recognition: A tutorial [J]. Proceedings of IEEE, 1997, 85(9): 1437-t462.
[4] B. Miller. Vital signs ofidentity [J]. IEEE Spectrum, 1994, 31(2): 22-30.
[5] V. Nalwa. Automatic on-line signature verification [J]. Proceedings of IEEE, 1997, 85(2): 213-239.
[6] R. Wildes. Iris Recognition: A emerging biometric technology [J]. Proceedings of IEEE, 1997, 85(9): 1348-1363.
[7] J. Zhang, Y. Yan, and M. Lades. Face Recognition: Eigenface, elastic matching and neural nets [J]. Proceedings of IEEE, 1997, 85(9): 1423-1436.
[8] R.L. Hsu, Mohamed Abdel-Mottaleb and A. K. Jain. Face detection in color images [J]. IEEE Transactions on Pattern Recognition and Machine Analysis, 2002, 24(5): 696-706.
[9] Anil K. Jain, Salil Parbhakar, Lin Hong, and Sharath Pankanti. Filterbank-Based Fingerprint Matching[J]. IEEE Transactions on Image Processing, 2000,9(5):846-859.
[10] J.L.Wayman. Fundamentals of Biometric Authentication Technologies [J]. International Journal of Image and Graphics, 2001, 1(1): 93-113.
[11] L. Hong, Y. Wan and A. K. Jain. Fingerprint image enhancement: algorithm and performance evaluation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998. 20(8): 777-789
[12] L.Hong,A.K.Jain, S.Pankanti and R.Bolle.Fingerprint enhancement[C].In: L. Abbott et al: Proceedings of 3rd IEEE Workshop on Applications of Computer Vision, IEEE Comp uter Society Press.Sarasota, Florida. 1996:202-207
[13] 尹义龙．詹小四，谭台哲，宁新宝．基于加博函数的指纹增强算法及应用[J]．软件学报，2003，14(3)：484-489
[14] 陈旭，尹义龙，王彦荣，王海洋．基于大窗口的频谱分析的指纹纹线距离的估计方法，复旦学报，2004，5：889-892
[15] Yi.Chen., Dass. S., Jain. A. Fingerprint quality indices for predicting authentication performance[J]. In: Proc. Audio-and Video-Based Biometne Person Authentication. Springer LNCS-3546.2005:160-170
[16] Klimanee C., Nguyen DT..On the Design of 2-D Gabor Filtering of Fingerprint Images [C].CD - Proceedings of 2004 IEEE Consumer Communications and Networking Conf erence (CCNC2004): Consumer Networking,Las Vegas,Nevada USA,2004
[17] Anil K. Jain- Salil Prabhakar, Lin Hong. A Multichannel Approach to Fingerprint Classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence;1999, 21(4):348-359
[18] Sen Wang, Yangsheng Wang. Fingerprint Enhancement in Singular Point Area, IEEE Signal Processing Letters. 2004.11(1):16-19
[19] Hong, L., Wan. Y., Jain, A. Fingerprint image enhancement: algorithms and performan ce evaluation. IEEE Transactions on Pattern Analysis and Machine Intelligence.1998,2 0:777—789
[20] Shen, L., Kot. A., Koo. W. Quality measures of fingerprint images. Audio- and Video-based Biometric Person Authentication. (2001)
[21] Lim E., Jiang, X., Yau. W. Fingerprint quality and validity analysis[C].IEEE International Conference on Image Processing, 2002(1):469-472
[22] Yun,Eun-Kyung, Cho.Sung-Bae.Adaptive fingerprim image enhancement with fingerprint image quality analysis[J]. 2006.24(1): 101-110
[23] Julian Fierrez-Aguilar, Yi Chen, Javier Ortega-Garcia, Anil Jain.Incorporating Image Quality in Multi-Algorithm Fingerprint Verification[J].Proc. ICB.Springer LNCS. 2006
[24] A. K. Jain. S. Prabhakar. and L. Hong, " A multichannel approach to fingerprint classification." IEEE Trans. Pattern Anal. Machine Intell., vol. 21. no. 4, pp. 348-359. 1999.
[25] N. Ratha, K. Karu. S. Chen and A. K. Jain. A Real-time Matching System for Large Fingerprint Database, IEEE Trans. on Patterm Anal. Machine Intell., Vol. 18, No.8, pp. 799-813, 1996.
[26] A. K. Jain, L. Hong and R. Bolle, On-line Fingerprint Verification, IEEE Trans. Pattern Anal. and Machine Intell., Vol. 19, No. 4, pp. 302-314, 1997.
[27] J. G. Daugman, "High confidence recognition of persons by a test of statistical independence," IEEE Trans. Pattern Anal. Machine Intell., vol. 15, no. 11, pp. 1148-1161, 1993.
[28] "Uncertainty, relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters," J. Opt. Soc. Amer. A, vol. 2, pp. 1160-1169, 1985.
[29] A. K. Jain, S. Prabhakar, and L. Hong, "A multichannel approach to fingerprint classification," IEEE Trans. Pattern Anal. Machine Intell., vol. 21, no. 4, pp. 348-359, 1999.
[30] 尹义龙．自动指纹识别系统研究：[博士论文]．长春：吉林工业大学，2000．
[31] T.Liu, G.C.Zhu, C.Zhang, P.W.Hao, Fingerprint Indexing Based on Singular Point Correlation, International Conference on Image Processing, Volume 3, 11-14 Sept. 2005 Page(s): 293-296
[32] Jain.A.K.Yi Chen.Demirkus. M. Pores and Ridges: High-Resolution Fingerprint Matching Using Level 3 Features, IEEE Trans. Pattern Anal. Machine Intell, Vol.29,no. 1,pp. 15-27,Jan. 2007
[33] R.K. Rowe, S.P. Corcoran. K.A. Nixon, and R. E. Ostrom, "Multi-spectral Imaging for Biometrics," Proc. SPIE Conf. Spectral Imaging: Instrumentation, Applications, and Analysis, pp. 90-99, Mar. 2005.
[34] A.K. Jain, Y. Chen, and M. Demirkus, "Pores and Ridges: Fingerprint Matching Using Level 3 Features," Proc. Int'l Conf. Pattern Recognition. vol. 4, pp. 477-480 Aug. 2006.
[35] D.Sherlock, D. M. Monro and K. Millard. Fingerprint enhancement by directional fourier filter. IEE Proceedings of Vision Image & Signal Processing, 1994, 141(2): 87～94.
[36] X. Xia and L. O'Gorman, "Innovations in Fingerprint Capture Devices," Pattern Recognition. vol. 36, no. 2, pp. 361-369, 2003.
[37] A. Ross, S.C. Dass, and A.K. Jain, "Fingerprint Warping Using Ridge Curve Correspondences," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 28, no. 1, pp. 19-30, Jan. 2006.
[38] S. Sangiorgi, A. Manelli, T. Congiu, A. Bini, G. Pilato, M. Reguzzoni, and M. Raspanti, "Microvascularization of the Human Digit as Studied by Corrosion Casting," J. Anatomy, vol. 204,pp. 123-131, 2004.
[39] D.C. Douglas Hung. Enhancement and feature purification of fingerprint images. Pattern Recognition, 1993, 26(11): 1661～1671.
[40] L.O'Gorman and J. V. Neckerson. An approach to fingerprint filter design. Pattern Recognition, 1989, 22(1): 29～38.
[41] G.T. Candela, P. J. Grother, C. I. Watson. R. A. Wilkinson, and C. L. Wilson. "PCASYS: A pattern-level classification automation system for fingerprints,", NIST Tech. Rep. NISTIR 5647, Aug. 1995.
[42] L. Hong and A. K. Jain, "Classification of fingerprint images," in 11th Scandinavian Conf. Image Analysis, Kangerlussuaq, Greenland, June 7-11, 1999.
[43] M. Eshera and K. S. Fu. "A similarity measure between attributed relational graphs for image analysis." in Proc. 7th Int. Conf. Pattern Recognition. Montreal. P.Q., Canada, July 30-Aug. 3 1984.
[44] S. Gold and A. Rangarajan. " A graduated assignment algorithm for graph matching," IEEE Trans. Pattern Anal.Machine Intell., vol. 18. no. 4, pp. 377-388, 1996.
[45] A. K. Hrechak and J. A. McHugh, "Automated fingerprint recognition using structural matching," Pattern Recognit., vol. 23. pp. 893-904. 1990.
[46] A. Ranade and A. Rosenfcld, "Point pattern matching by relaxation." Pattern Recognit., vol. 12. no. 2. pp. 269-275. 1993.
[47] J. Ton and A. K. Jain. "Registering landsat images by point matching," IEEE Trans. Geosci. Remote Sensing, vol. 27. pp. 642-651. May 1989.
[48] E. C. Driscoll. C. O. Martin. K. Ruby. J. J. Russel. and J. G. Watson, "Method and apparatus for verifying identity using image correlation." U.S. Patent 5 067 162, 1991.
[49] A. Sibbald. "Method and apparatus for fingerprint characterization and recognition using auto-correlation pattern." U.S. Patent 5 633 947. 1994.
[50] A. K. Jain. L. Hong. S. Pankanti. and R. Bolle, " An identity authentication system using fingerprints. " Proc. IEEE. vol. 85. pp. 1365-1388,Sept. 1997.