臺灣博碩士論文加值系統

近年來，奇異點(含核心點與三角點)偵測演算法已被廣泛的應用在指紋分類及辨識上。為解決偵測上的問題，本論文提出了一個新潁的奇異點偵測演算法，利用適應性的影像增強技術、簡潔的邊界切割、以及採用不可分離的二維小波極值點以用於定位。針對低對比度的影像，我們特別使用經適應性處理後的等化影像來取代原始影像的亮度資訊以提高對比度。經由二值化的處理後，我們從影像兩邊開始搜尋定位地標點，根據定位地標點再使用多邊型來切割出感興趣指紋區塊(IOI)。接著採用彭佳勒指標(Poincare index)演算法偵測出奇異點，並將指紋影像以核心點為中心進行轉正。最後，將彭佳勒指標所偵測到的核心點位置進一步搜尋附近之曲率變化最大的點，奇異點之標準位置則是以Henry所定義之奇異點位置為主。
我們測試FVC2002 DB1與DB2兩個指紋資料庫，每個資料庫分別有800張指紋影像。由實驗結果得知DB1資料庫平均正確接受率為89.5%而核心點的平均錯誤接受率則是0.4%; DB2資料庫平均正確接受率為88%而核心點的平均錯誤接受率則是0.43%。此外，我們同時與其他文獻現有技術中最好的結果相比較，本研究之正確接受率則表現較佳。

The singular points, core and delta, are widely used in fingerprint classification and identification. In this thesis, a new singular point detection algorithm based on adaptive image enhancement, compact boundary segmentation, and 2-D non-separable wavelet extrema for localization is proposed. We perform adaptive correction for low contrast image to replace the intensity information of a given image matrix with another equalized intensity matrix of the image. With the help of binary thresholding, the impression-of-interest (IOI) region is segmented via the landmarks detected by searching and starting from both sides of the image and framed with the connected polygons. Following the orientation filed estimation and smoothing and the computation of Poincare Index, the fingerprint IOI is aligned per the singular points and the orientation filed nearby. Finally, the location of the detected core is further refined by searching the neighboring curve with the largest curvature based on Henry system. The proposed framework has been tested on the FVC2002 DB1 and DB2 databases. For the 800 images in the each database set, DB1 correctly detection rate reaches 89.5% with core false alarm rate 0.4% in average, while DB2 correctly detection rate reaches 88% with core false alarm rate 0.43% in average. The detection rates are compared favorably to the state-of-the-art singular point detectors.

Chinese abstract ............................................................................................... i
English abstract .............................................................................................. iii
Gratitude ......................................................................................................... v
Contents ......................................................................................................... vi
Table of list ................................................................................................... vii
Figure of list ................................................................................................ viii
Chapter 1 Preface .......................................................................................... 1
1.1 Introduction ................................................................................................ 1
1.2 Related works ............................................................................................ 3
1.3 Motivation and contribution .................................................................... 4
1.4 Thesis organization ................................................................................... 4
Chapter 2 Methodology Analysis ................................................................... 9
2.1 Background removal using Singular Value Decomposition (SVD) ............. 9
2.2 Energy transformation ............................................................................ 18
2.3 Fingerprint contour ................................................................................. 23
2.4 Blur detection using 2-D non-separable wavelet entropy filtering ............ 27
2.5 Wavelet Extrema for Singular Point Detection (WESPD) ................ 31
Chapter 3 Singular Point detection ............................................................. 39
3.1 Orientation field estimation ................................................................... 42
3.2 Image enhancement ................................................................................. 44
3.3 Hierarchical singular points detection using Poincare Index method ........ 47
Chapter 4 Experiment Results and Discussion ........................................... 49
Chapter 5 Conclusion and Future Work ..................................................... 63
References .................................................................................................... 64


Table of list
Table 1 Calculated mean and standard deviation in each database for
synthetic Gaussian distribution function ………………………… 17
Table 2 Entropy of SVD-enhanced images compared with grayscale
images for each database ……………………………………… 18
Table 3 The comparison results of different detection algorithm on
FVC02’s DB1 ………………………… 53
Table 4 The comparison results of different detection algorithm on
FVC02’s DB2 …………………………53
Table 5 The comparison results of different detection algorithm on NIST
4 database ……………………………… 54


Figure of list
Fig. 1. 8-class fingerprints displayed with core in red color and delta in green color ............................................7
Fig. 2. Ground truth examples of core points based on the Henry system ........... 8
Fig. 3. Fingerprint images in FVC02 DB1 .......................... 13
Fig. 4. Fingerprint images in FVC02 DB2 .......................... 14
Fig. 5. Background removal of Fig. 3 by using SVD enhancement .... 15
Fig. 6. Background removal of Fig. 4 by using SVD enhancement .... 16
Fig. 7. Result of Fig. 3 by using energy transformation .......... 21
Fig. 8. Result of Fig. 4 by using energy transformation .......... 22
Fig. 9. Two-way horizontal projections for landmark detection .... 24
Fig. 10. Boundary segmentation of Fig. 3 ................... 25
Fig. 11. Boundary segmentation of Fig. 4 ................... 26
Fig. 12. Filter bank implementation of 2-D nonseparable wavelet transform ..................................................... 28
Fig. 13. Blur detection result by 2-D non-separable wavelet entropy
filtering for high quality images ............................. 29
Fig. 14. Blur detection result by 2-D non-separable wavelet entropy
filtering for low quality images .............................. 30
Fig. 15. Singular point detection flowchart ................... 31
Fig. 16. Image alignment ...................................... 35
Fig. 17. Skeletonized sub-region and ridge curves ............. 36
Fig. 18. Wavelet extrema in the subregion ..................... 36
Fig. 19. Two 8-adjacency grids moving towards each other along the
ridge curve curved in yellow .................................. 37
Fig. 20. Traced path of the ridge curve ....................... 37
Fig. 21. The singular point located at the lowest ridge curve and the
underneath area ............................................ 38
Fig. 22. Detection of singular point in accordance of the Henry system ..................................................... 38
Fig. 23. System flowchart of the proposed method ........... 41
Fig. 24. Examples of orientation filed estimation .......... 43
Fig. 25. Ridge enhancement using Gabor filtering ........... 46
Fig. 26. Comparison results of singular point detection .... 56
Fig. 27. We perform results of singular point detection in FVC02 DB1 ........................................................ 56
Fig. 28. Results of singular points truly detection on FVC02 DB2 .......................................................... 57
Fig. 29. Results of singular points truly detection on NIST 4 database ................................................ 58
Fig. 30. Low quality results of singular points detection ....... 59
Fig. 31. Results of fingerprint ..................... 59
Fig. 32. False result of singular points false detection on FVC02 DB1 ..... 60
Fig. 33. False result of singular points false detection on FVC02 DB2 ..... 61
Fig. 34. Miss result of singular points false detection on FVC02 DB2 ..... 62
Fig. 35. Results of harm fingerprint image.................. 62

[1]C. F. Hsu and J. W. Wang, “Fingerprint classification based on hierarchical singular point detection and traced orientation flow,” CVGIP2008, I-Lan County, Taiwan, Aug. 24-26, 2008.
[2]D. Maltoni, D. Maio, A. K. Jain, and S. Probhaker, Handbook of Fingerprint Recognition, Springer-Verlag, 2003.
[3]R. E. Henry, Classification and used of finger prints, George Rutledge & Sons, London, 1900.
[4]K. Karu and A. K. Jain, “Fingerprint classification,” Pattern Recognition, vol. 29, no. 3, pp.389-404, 1996.
[5]A. K. Jain, S. Prabhakar, and L. Hong, “A multichannel approach to fingerprint classification,” IEEE Trans. on Pattern Analysis and Machine Intell., vol. 21, no. 4, pp. 348-359, 1999.
[6]A. K. Jain, S. Prabhakar, L. Hong, and S. Pankanti, “Filterbank-based fingerprint matching,” IEEE Trans. on Image Processing, vol. 9, no. 5, pp. 846 - 859, 2000.
[7]A. Senior, “A combination fingerprint classifier,” IEEE Trans. on Pattern Anal. Machine Intell., vol. 23, no. 10, pp. 1165-1174, 2001.
[8]A. Ahmadyfard and M. S. Nosrati, “A novel approach for fingerprint singular points detection using 2-D-wavelet,” IEEE Computer Systems and Applications, pp. 688-691, 2007.
[9]A. P. Fitz and R. J. Green, “Fingerprint classification using a hexagonal fast Fourier transform,” Pattern Recognition, vol. 29, no. 10, pp. 1587-1597, 1996.
[10]C. J. Lee, I. H. Jeng, T. N. Yang, C. J. Chen, and K. L. Lin, “Singular points detection in fingerprint images using Gabor transform,” 9th International Conference on Signal Processing, pp. 2078-2081, 2008.
[11]H. Demirel and G. Anbarjafari, “Pose invariant face recognition using probability distribution function in different color channels,” IEEE Signal Processing Letters, vol. 15, pp 537 - 540, 2008.
[12]M. Unser, “Texture classification and segmentation using wavelet frames,” IEEE Trans. on Image Processing, vol. 4, no. 11, pp. 1546-1560, 1995.
[13]R. A. Gopinath and C. S. Burrus, “Oversampling invariance of wavelet frames”, IEEE-SP Int. Symposium on Time-Frequency and Time-Scale Analysis, pp. 375-378,1992.
[14]S. Mallat, A Wavelet Tour of Signal Processing, Third Ed., Associated Press, 2009.
[15]W. Miao, L. Yan, P. Quan, and Z. Hong, “A Gabor filter based fingerprint enhancement algorithm in wavelet domain,” IEEE Int. Symposium on Communications and Information Technology, vol. 2, pp. 1468-1471, 2005.
[16]A. K. Jain, L. Hong, and R. Bolle, “On-line fingerprint verification”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 9, pp. 302 - 314, 1997.
[17]J. Yang, L. Liu, T. Jiang, and Y. Fan, “A modified Gabor filter design method for fingerprint image enhancement,” Pattern Recognition, vol. 24, pp. 1805-1817, 2003.
[18]J. Liu, W. Li, and Y. Tian, “Automatic thresholding of gray-level pictures using two-dimension Otsu method,” International Conference on Circuits and Systems, vol. 1, pp. 325-327, 1991.
[19]J. Soo and K. Hyeon, “Palmprint identification algorithm using invariant moments and Otsu binarization,” ACIS International Conference on Computer and Information Science, pp. 97-99, 2005.
[20]J. Li, T. Lei, W. Rong, G. Lie, and C. Jiang, “An improved Otsu image segmentation algorithm for path mark detection under variable illumination,” IEEE Proceedings, pp. 840-844, 2005.
[21]http://paginas.fe.up.pt/~spd2010/
[22]http://bias.csr.unibo.it/fvc2002
[23]M. Tico and P. Kuosmanen, “A multiresolution method for singular points detection in fingerprint images,” IEEE Int. Symposium on Circuit and System, vol. 4, pp. 183-186, 1999.
[24]P. Ramo, M Tico, V. Onnia, and J. Saarinen, “Optimized singular point detection algorithm for fingerprint images,” IEEE Trans. on Image Processing, vol. 3, no. 3, pp. 242-245, 2001.
[25]J. Zhou, F. Chen, and J. Gu, “A novel algorithm for detecting singular points from fingerprint images,” IEEE Trans. Pattern Anal. And Machine Intell., vol. 31, no. 7, pp. 1239-1250, July 2009.
[26]S. Chikkerur and N. K. Ratha, “Impact of singular point detection on fingerprint matching performance,” Proc. Fourth IEEE Workshop on Automatic Identification Advanced Technologies, pp. 207-212, 2005.