過乾、過濕的指紋與指紋品質有著密切關係，乾指紋上的突脊(ridge)碎裂且不連續，而濕指紋造成突脊模糊、黏在一塊，且汗毛孔被水氣蓋住，這些現象造成指紋特徵擷取不易，使得指紋辨識率降低。為了判斷由乾溼問題所造成的低品質指紋，我們提出了一套乾濕指紋的分類方法，利用指紋突脊的面積占比、平滑性、連續性以及突脊上的雜質數量當作分辨乾濕指紋的特徵，並且量化每一個特徵，最後將這些特徵分數利用支持向量機(SVM)分類。由於我們採用指紋突脊的特徵，而這些特徵可以在不同的解析度中呈現出來，因此針對不同解析度的指紋都可以精準的分類。實驗中，我們也利用FVC 2002 DB1(500dpi)與NASIC Database(2,000dpi)兩種不同的解析度來驗證我們的方法。另外，我們也比較了Wu與 Awasthi的方法，結果顯示我們的方法優於以上兩者，且我們的方法與Wu的方法合併使用時，可以獲得更佳的分類正確率。在FVC 2002 Db1中乾、正常、濕指紋的正確率為98.31%、96.43%、95.03%，以及NASIC Database中乾、正常、濕指紋的正確率是96.00%、95.43%、100%。最後，我們也證明利用我們的品質分類方法，排除FVC2002 Db1中指紋的乾溼區域後再進行指紋配對可以降低EER，由10.39降低至9.79，降低幅度為5.68%。

The dry and wet fingerprint images affect the image quality. The ridges on dry fingerprints are broken and discontinuity. The ridges on wet fingerprints are thick or stick together. These phenomena cause feature extraction difficult and lead to decrease recognition accuracy. In order to distinguish low quality image caused by dry and wet fingerprint, we design a dry and wet fingerprint image classification method. We propose several novel ridge features, ridge region ratio, ridge smoothness, ridge continuity, and ridge impurity. Then,we quantify these features and use SVM to divide fingerprint images into three categories, dry, normal and wet. Because these adopted ridge features can be observed clearly on different fingerprint image resolutions, the proposed method works accurately for multiple fingerprint image resolutions.

We verify the proposed method on FVC 2002 Db1 (500dpi) and NASIC database (2,000dpi). The SVM classification accuracy are 98.31%、96.43%、95.06% on the FVC2002 DB1, and 96.00%、95.43%、100% on the NASIC database for dry, normal and wet respectively. In addition, we also prove that excluding the wet and dry region on a whole fingerprint can improve recognition result. The equal error rate is reduced from 10.39% to 9.79% for FVC2002 Db1. The improvement rate is 5.68%.

1 Introduction 1
1.1 Motivation and Problem Description . . . . . . . . . . . . . . . . . 1
1.2 Contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 Related Work 9
2.1 Local Texture Based . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Ridge-Valley Based . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Hybrid Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3 Ridge Feature Extraction for Multiple Fingerprint Resolutions
and Classifications 15
3.1 Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1.1 Contrast Limited Adaptive Histogram Equalization . . . . . 16
3.1.2 Otsu’s Method . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2 Ridge Feature Extraction . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.1 Ridge Region Ratio . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.2 Ridge Smoothness . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.3 Ridge Continuity . . . . . . . . . . . . . . . . . . . . . . . . 26
i
3.2.4 Ridge Impurity . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.5 Ridge Period . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4 Experimental Result 33
4.1 Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.1.1 NASIC Database . . . . . . . . . . . . . . . . . . . . . . . . 34
4.1.2 FVC2002 Database . . . . . . . . . . . . . . . . . . . . . . . 36
4.2 Classification Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3 Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.4 Verify with a Fingerprint Recognition System . . . . . . . . . . . . 46
4.4.1 Data Preparation . . . . . . . . . . . . . . . . . . . . . . . . 46
4.4.2 Classification Mask . . . . . . . . . . . . . . . . . . . . . . . 48
4.4.3 Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5 Conclusion 52

[1] A. Awasthi, K. Venkataramani, and A. Nandini, “Image quality quantification
for fingerprints using quality-impairment assessment,” in Applications
of Computer Vision (WACV), 2013 IEEE Workshop on. IEEE, 2013, pp.
296–302.
[2] F. Alonso-Fernandez, J. Fierrez, J. Ortega-Garcia, J. Gonzalez-Rodriguez,
H. Fronthaler, K. Kollreider, and J. Bigun, “A comparative study of fingerprint
image-quality estimation methods,” IEEE Transactions on Information
Forensics and Security, vol. 2, no. 4, pp. 734–743, 2007.
[3] C. Wu, S. Tulyakov, and V. Govindaraju, “Image quality measures for fingerprint
image enhancement,” Multimedia Content Representation, Classification
and Security, pp. 215–222, 2006.
[4] E. Tabassi and C. L. Wilson, “A novel approach to fingerprint image quality,”
in Image Processing, 2005. ICIP 2005. IEEE International Conference on,
vol. 2. IEEE, 2005, pp. II–37.
[5] E. Lim, X. Jiang, and W. Yau, “Fingerprint quality and validity analysis,”
in Image Processing. 2002. Proceedings. 2002 International Conference on,
vol. 1. IEEE, 2002, pp. I–I.
[6] M. A. Olsen, H. Xu, and C. Busch, “Gabor filters as candidate quality measure
for nfiq 2.0,” in Biometrics (ICB), 2012 5th IAPR International Conference
on. IEEE, 2012, pp. 158–163.
[7] Y. Zhao, C. Jiang, X. Fang, and B. Huang, “Research of fingerprint image
quality estimation,” in Dependable, Autonomic and Secure Computing, 2009.
DASC’09. Eighth IEEE International Conference on. IEEE, 2009, pp. 791–
795.
[8] Y. Chen, S. C. Dass, and A. K. Jain, “Fingerprint quality indices for predicting
authentication performance,” in AVBPA, vol. 3546. Springer, 2005, pp.
160–170.
[9] S. Joun, H. Kim, Y. Chung, and D. Ahn, “An experimental study on measuring
image quality of infant fingerprints,” in Knowledge-Based Intelligent
Information and Engineering Systems. Springer, 2003, pp. 1261–1269.
[10] Z. Shi, Y. Wang, J. Qi, and K. Xu, “A new segmentation algorithm for low
quality fingerprint image,” in Image and Graphics (ICIG’04), Third International
Conference on. IEEE, 2004, pp. 314–317.
[11] Q. Zhao, F. Liu, L. Zhang, and D. Zhang, “A comparative study on quality
assessment of high resolution fingerprint images,” in Image Processing (ICIP),
2010 17th IEEE International Conference on. IEEE, 2010, pp. 3089–3092.
[12] P. Grother and E. Tabassi, “Performance of biometric quality measures,”
IEEE transactions on pattern analysis and machine intelligence, vol. 29, no. 4,
pp. 531–543, 2007.
[13] T. P. Chen, X. Jiang, and W.-Y. Yau, “Fingerprint image quality analysis,”
in Image Processing, 2004. ICIP’04. 2004 International Conference on, vol. 2.
IEEE, 2004, pp. 1253–1256.
[14] J. Amengual, A. Juan, J. Pérez, F. Prat, S. Sáez, and J. Vilar, “Real-time
minutiae extraction in fingerprint images,” 1997.
[15] A. K. Jain, Y. Chen, and M. Demirkus, “Pores and ridges: High-resolution
fingerprint matching using level 3 features,” IEEE Transactions on Pattern
Analysis and Machine Intelligence, vol. 29, no. 1, pp. 15–27, 2007.
[16] Q. Zhao, L. Zhang, D. Zhang, N. Luo, and J. Bao, “Adaptive pore model for
fingerprint pore extraction,” in Pattern Recognition, 2008. ICPR 2008. 19th
International Conference on. IEEE, 2008, pp. 1–4.
[17] L. Shen, A. Kot, and W. M. Koo, “Quality measures of fingerprint images,”
in AVBPA. Springer, 2001, pp. 266–271.
[18] R. Syam, M. Hariadi, and M. H. Purnomo, “Determining the dry parameter
of fingerprint image using clarity score and ridge-valley thickness ratio.”
[19] L. Hong, Y. Wan, and A. Jain, “Fingerprint image enhancement: Algorithm
and performance evaluation,” IEEE transactions on pattern analysis and machine
intelligence, vol. 20, no. 8, pp. 777–789, 1998.
[20] C. Jain-Cong and L. Shang-Hong, “Defective region detection in fingerprint
images with fully convolutional network.”
[21] M. A. Olsen, M. Dusio, and C. Busch, “Fingerprint skin moisture impact on
biometric performance,” in Biometrics and Forensics (IWBF), 2015 International
Workshop on. IEEE, 2015, pp. 1–6.
[22] E. D. Pisano, S. Zong, B. M. Hemminger, M. DeLuca, R. E. Johnston,
K. Muller, M. P. Braeuning, and S. M. Pizer, “Contrast limited adaptive
histogram equalization image processing to improve the detection of simulated
spiculations in dense mammograms,” Journal of Digital imaging, vol. 11,
no. 4, pp. 193–200, 1998.
[23] N. Otsu, “A threshold selection method from gray-level histograms,” IEEE
transactions on systems, man, and cybernetics, vol. 9, no. 1, pp. 62–66, 1979.
[24] G. Rafael and W. Richard, Digital image processing, 3rd ed. Pearson, 2009.
[25] J. Kittler, “On the accuracy of the sobel edge detector,” Image and Vision
Computing, vol. 1, no. 1, pp. 37–42, 1983.
[26] J. Shi et al., “Good features to track,” in Computer Vision and Pattern Recognition,
1994. Proceedings CVPR’94., 1994 IEEE Computer Society Conference
on. IEEE, 1994, pp. 593–600.
[27] V. Vapnik, The nature of statistical learning theory. Springer science &
business media, 2013.
[28] C.-C. Chang and C.-J. Lin, “Libsvm: a library for support vector machines,”
ACM transactions on intelligent systems and technology (TIST), vol. 2, no. 3,
p. 27, 2011.
[29] D. Maio, D. Maltoni, R. Cappelli, J. L. Wayman, and A. K. Jain, “Fvc2002:
Second fingerprint verification competition,” in Pattern recognition, 2002.
Proceedings. 16th international conference on, vol. 3. IEEE, 2002, pp. 811–
814.
[30] T.-T. Chu and C.-T. Chiu, “A cost-effective minutiae disk code for fingerprint
recognition and its implementation,” in Acoustics, Speech and Signal Processing
(ICASSP), 2016 IEEE International Conference on. IEEE, 2016, pp.
981–985.
[31] A. K. Jain and F. Farrokhnia, “Unsupervised texture segmentation using gabor
filters,” in Systems, Man and Cybernetics, 1990. Conference Proceedings.,
IEEE International Conference on. IEEE, 1990, pp. 14–19.