臺灣博碩士論文加值系統

近幾年使用個人生物特徵來辨識身份已逐漸受到人們的重視。而由於使用上的便利性及辨識正確率等因素，虹膜辨識技術在眾多的生物特徵辨識技術中已脫穎而出，廣為各界所採用。但傳統的虹膜辨識系統是假設使用者必須完全配合系統取像，以便攝得正視之虹膜影像。此假設條件也限制了虹膜辨識系統的應用層面，在更廣泛的應用中，虹膜辨識系統必須也要能處理非正向視角 (Non-orthogonal View) 所取得的虹膜影像。本論文將深入探討使用非正向視角的虹膜影像以辨識個人身份的技術。新的取像系統包含一個雙 CCD 攝影機與可移動式的螢幕，以取得不同視角的紅外線灰階及可見光彩色之四頻譜虹膜影像。四頻譜虹膜影像所提供的豐富資訊可以簡化虹膜區域的定位。為了比對由不同視角所攝得的虹膜影像，我們提出仿射轉換 (Affine Transform) 以更正虹膜特徵在不同視角下的的變形。且仿射轉換的參數可經由瞳孔的邊界所求得。因此，四頻譜虹膜影像經由仿射轉換矯正，以將非正向視角所造成的影響降至最低。非正向視角的取像較無限制，所以傳統的虹膜區域的定位法失敗率相當高。所以，我們利用四頻譜虹膜影像所提供的資訊，發展一個更強健的虹膜區域定位的方法。經由實驗證實，轉換過的非正向視角虹膜影像與正向視角的虹膜影像相似度高，因此傳統的特徵萃取方法即可適用於此並且達到良好的辨識率。

In recent years, techniques for personal identification with biometrics have drawn more and more attention from people in the world. Because of the user friendliness and the high recognition accuracy, iris recognition techniques have stood out among the biometric personal identification techniques and are widely adopted in many applications. However, the traditional iris recognition systems are developed under the assumption that the user has to be cooperative in order for the system to acquire an orthogonal view iris image. This restriction has reduced the applicability of iris recognition systems. An iris recognition technique must also be able to deal the non-orthogonal view iris image to apply the iris recognition techniques in more extensive applications. In this thesis, we will discuss in depth an iris recognition technique that accepts the non-orthogonal view iris image for personal identification. A new imaging system containing a dual CCD camera and a relocatable LCD monitor is described which can provide four-spectral iris images taken at different off-axis angles. A four-spectral iris image contains plentiful information which can be used to simplify the iris localization task. In order to match iris images acquired with different off-axis angles, we propose to model the iris pattern deformation with an affine transformation. The transformation parameters can be estimated with the pupillary boundary. Thus, the four-spectral iris image can be rectified to minimize the effect of a non-zero off-axis angle. Because the iris localization problem of a non-orthogonal view iris image is more difficult than that of an orthogonal one, the traditional iris localization method will cause very high failure rate in processing the non-orthogonal view iris image. Therefore, we developed a new robust method for detecting iris boundaries using four spectral images. Real experiments have been conducted to show that the non-orthogonal view iris image can be converted to fit the orthogonal view imaging condition and thus the traditional feature extraction methods can be used to achieve high recognition rate.

1 論文簡介
1.1 研究動機. . . . . . . . . . . . . . . . . . . . . . 1
1.2 文獻探討. . . . . . . . . . . . . . . . . . . . . . 3
1.3 論文具體貢獻 . . . . . . . . . . . . . . . . . . . . 4
1.4 論文結構. . . . . . . . . . . . . . . . . . . . . . 5
2 系統硬體架構與計算流程
2.1 簡介. . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 系統硬體架構 . . . . . . . . . . . . . . . . . . . . 6
2.2.1 雙CCD 攝影機影像校正. . . . . . . . . . . . . . . . 9
2.3 軟體流程. . . . . . . . . . . . . . . . . . . . . . 12
3 四頻譜虹膜影像定位
3.1 簡介. . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 瞳孔定位. . . . . . . . . . . . . . . . . . . . . . 15
3.2.1 交錯偵測與活體偵測. . . . . . . . . . . . . . . . . 16
3.3 仿射矯正. . . . . . . . . . . . . . . . . . . . . . 17
3.4 虹膜定位. . . . . . . . . . . . . . . . . . . . . . 18
3.5 眼瞼定位. . . . . . . . . . . . . . . . . . . . . . 20
4 正規化虹膜影像與特徵萃取
4.1 簡介. . . . . . . . . . . . . . . . . . . . . . . . 22
4.2 正規化虹膜影像 . . . . . . . . . . . . . . . . . . . 22
4.3 特徵萃取. . . . . . . . . . . . . . . . . . . . . . 24
4.4 正規化漢明距 . . . . . . . . . . . . . . . . . . . . 24
5 實驗結果
5.1 簡介. . . . . . . . . . . . . . . . . . . . . . . . 26
5.2 實驗環境. . . . . . . . . . . . . . . . . . . . . . 26
5.3 辨識結果參數說明. . . . . . . . . . . . . . . . . . . 26
5.4 實驗結果. . . . . . . . . . . . . . . . . . . . . . 27
6 結論與未來展望
6.1 結論. . . . . . . . . . . . . . . . . . . . . . . . 34
6.2 未來展望. . . . . . . . . . . . . . . . . . . . . . 34

[1] 吳庭榕, “多視角虹膜辨識系統,” 國立暨南國際大學, 2005.
[2] J. Daugman, “High confidence visual recognition of persons by a test of statistical independence,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, pp. 1148–1161, November 1993.
[3] R. Wildes, “Iris recognition:an emerging biometric technology,” Proceedings of the IEEE, vol. 85, pp. 1348–1363, September 1997.
[4] J. Daugman and C. Downing, “Recognizing iris texture by phase demodulation,” Image Processing for Biometric Measurement, IEE Colloquium on, pp. 1–8, Apr 1994.
[5] J. Daugman, “Biometric personal identification system based on iris analysis,” United States Patent, Patent Number:5291560, IriScan Inc., March 1994.
[6] J. Daugman, “The importance of being random: Statistical principles of iris recognition,” Pattern Recognition, vol. 36, no. 2, pp. 279–291, 2001.
[7] J. Daugman and C. Downing, “Epigenetic randomness, complexity and singularity of human iris patterns,” Proceedings of the Royal Society, B. 268, Biological Sciences, pp. 1737–1740, 2001.
[8] R. Wildes, J. Asmuth, G. Green, and S. Hsu, “A system for automated iris recognition,” Proceedings of the IEEE Workshop on Applications of Computer Vision, pp. 121–128, 1994.
[9] R.Wildes, J. Asmuth, K. Hanna, S. Hsu, R. Kolczynski, J. Matey, and S. McBride,“Automated, non-invasive iris recognition system and method,” United States Patent, Patent Number:5572596, David Sarnoff Research Center, Inc., November 1996.
[10] L.Ma, T. Tan, Y.Wang, and D. Zhang, “Efficient iris recognition by characterizing key local variations,” IEEE Transactions on Image Processing, vol. 13, pp. 739–750, June 2004.
[11] S. Lim, K. Lee, O. Byeon, and T. Kim, “Efficient iris recognition through improvement of feature vector and classifier,” Electronics and Telecommunications Research Institute(ETRI), vol. 23, pp. 61–70, June 2001.
[12] W. Boles and B. Boashash, “A human identification technique using images of the iris and wavelet transform,” IEEE Transactions on Signal Processing, vol. 46, pp. 1185–1198, April 1998.
[13] J. Huang, Y. Wang, T. Tan, and J. Cui, “A new iris segmentation method for recognition,” Proceedings of the 17th International Conference on Pattern recognition (ICPR), 2004.
[14] W. Kong and D. Zhang, “Accurate iris segmentation based on novel reflection and eyelash detection model,” Proceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing, May 2001.
[15] Z. Sun, Y. Wang, T. Tan, and J. Cui, “Improving iris recognition accuracy via cascaded classifiers,” Systems, Man and Cybernetics, Part C, IEEE Transactions on, vol. 35, pp. 435–441, August 2005.
[16] R. Gaunt, B. Bonney, R. Ives, D. Etter, and R. Schultz, “Collection and segmentation of non-orthogonal irises for iris recognition,” 2005.
[17] B. Bonney, R. W. Ives, D. M. Etter, and Y. Du, “Iris pattern extraction using bit-plane analysis,” in Proceedings of the IEEE 38th Annual Asilomar Conference on Signals, Systems, and Computers, vol. 1, pp. 582–586, 2004.
[18] V. Dorairaj, N. A. Schmid, and G. Fahmy, “Performance evaluation of non-ideal iris based recognition system implementing global ica encoding,” in Proceedings of the IEEE International Conference on Image Processing, vol. 3, pp. 285–288, 2005.
[19] A. Abhyankar, L. Hornak, and S. Schuckers, “Off-angle iris recognition using biorthogonal wavelet network system,” in Proceedings of the IEEE Workshop on Automatic Identification Advanced Technologies, pp. 239–244, 2005.
[20] H. Proen¸ca and L. Alexandre, “Ubiris: A noisy iris image database,” 2005. http://iris.di.ubi.pt/.
[21] CASIA Iris Image Database, http://www.sinobiometrics.com/.
[22] A. Fitzgibbon, M. Pilu, and R. B. Fisher, “Direct least square fitting of ellipse,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, no. 5, pp. 476–480, 1999.
[23] J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Machine Intell., vol. 8, no. 6, pp. 679–714, 1986.
[24] R.-E. Fan, P.-H. Chen, and C.-J. Lin, “Working set selection using second order information for training SVM,” Journal of Machine Learning Research, vol. 6, pp. 1889–1918, 2005.
[25] M. A. Fischler and R. C. Bolles, “Random sample consensus, a paradigm for model fitting with applications to image analysis and automated cartography,” Communications of the ACM, vol. 24, no. 6, pp. 381–395, 1981.
[26] J. Daugman, “How iris recognition works,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 14, no. 1, pp. 21–30, 2002.
[27] C.-T. Chou, S.-W. Shih, W.-S. Chen, and V. W. Cheng, “Iris recognition with multi-scale edge-type matching,” in the Proceedings of the IAPR International Conference on Pattern Recognition, to appear, 2006.