臺灣博碩士論文加值系統

隨著科技的發展進步，利用偽造的身份犯案的事件層出不窮，使得正確的身份認證變成十分受到重視的技術，例如：在汽車、筆記型電腦上加入指紋辨識系統，而且，手機市場上也首度出現了指紋辨識手機。由於恐怖份子的攻擊，使得許多國際機場開始裝設了虹膜、指紋等身份辨識系統，顯示出生物認證的重要。生物統計學是藉由人類個體特有的生理和行為特徵，進行身分識別和(或)個體驗證的一門科學；所以，愈來愈多的識別系統是以生物統計學為基礎作為發展的方向。其中，又以虹膜最具有獨一無二的特徵。
傅立葉轉換的特性之一，即是其在空間域的位移，並不會影響頻率域的資訊。我們從傅立葉轉換所獲得的頻率資訊是全域性的，因此，如原始信號在區域性所發生的變化情形，則會影響傅立葉轉換後的係數，這是因為傅立葉轉換不具備多重解析度的特性，因此，我們需要一個更佳的轉換。
過去幾年來，函數對於區域性頻率的分析已運用得非常普遍，尤其對低頻部份具長時間的解析度，與對高頻部份具有短時間的解析度；但是，於原始信號內產生一點小位移，將會影響小波係數，此乃是小波函數不能廣泛運用於圖型識別的原因。基於傅立葉轉換與小波轉換各具有其特性，因此在本論文中，我們希望結合此兩種轉換的特性，將其運用在虹膜辨識上，並建立一套自動辨識系統。

With the development and progress of science and technology, the incident of utilizing the counterfeit identity to commit a crime emerges more and more. Making correct identity authentication turns into an important technology. For example, join fingerprint recognition system on the automobile, on the notebook, and iris recognition system for the first time on the cell-phone. Due to the attack from the terrorism, lots of International airports begin to install the identities, such as iris, fingerprint, etc. Biometric is a science of research for physiological and behavioral characteristics. Therefore, more identification system is based on Biometric as an aspect of development. Among them, iris has the unique features.
A property of the Fourier transform is shift-invariant, that is, the translation in spatial domain cannot to affect the information of the frequency domain. However it is global information from the Fourier transform, the condition of the local variant can change the Fourier coefficient. Further the Fourier transform do not has the property of multi-resolution. Therefore, we need to develop the better transform.
In past few years, the Wavelet function is wildly used to analysis the local frequency. Specially, it is for low frequency has long-time resolution and high frequency has short-time resolution. But the Wavelet coefficient will be changed by a little translation in the original signal; this is why the Wavelet function cannot use for pattern recognition. Due to the special own property of the Fourier transform and Wavelet function, we expect to combine the property of these two transforms, and apply to iris recognition to build an automatic identify system.

誌謝
摘要
ABSTRACT
目錄
表目錄
圖目錄
1.緒論
1.1 研究動機
1.2 章節架構
2.文獻探討
2.1 影像前處理
2.1.1 虹膜定位
2.1.2 虹膜正規化
2.1.3 影像強化
2.2 特徵萃取
2.3 不變性
2.4 特徵比對
3.理論基礎
3.1 傅立葉轉換
3.2 小波轉換
3.3 統計的決策理論
3.3.1 Neyman-Pearson定理
3.3.2 接收者操作特性曲線(ROC Curve)
4.研究方法
4.1 虹膜影像定位
4.1.1 瞳孔中心定位
4.1.2 虹膜內外輪廓定位
4.1.3 虹膜正規化
4.1.4 特徵強化
4.2 虹膜特徵萃取
4.2.1 一維傅立葉轉換
4.2.2 一維小波轉換
4.2.3 特徵向量
4.3 特徵比對
5.虹膜辨識實驗
5.1 虹膜影像資料庫
5.2 實驗一：一維傅立葉轉換之評估
5.3 實驗二：使用各種一維小波濾波器之評估
5.4 實驗三：特徵向量執行小波分解次數之評估
5.5 實驗四：不同相似度比對方式對辨識結果之評估
5.6 實驗五：使用FFT-WT方法與高斯-赫米特動差之比較
5.7 討論
5.7.1 虹膜資料庫的數量增加
5.7.2 辨識率的提高
5.7.3 辨識速度的提昇
6.結論
參考文獻

[1] Jain, A., Bolle, R., and Pankanti, S., Eds, Biometrics, Personal Identification in Networked Society, Kluwer Academic Publishers, London, Vol. 479, Jan., 1999.
[2] Miller, B., “Vital Signs of Identity,” IEEE Spectrum, Vol. 31, pp. 22-30, 1994.
[3] Daugman, J. G., “How Iris Recognition Works,” IEEE Transactions on Circuits and Systems for Video Technology, pp. 21-30, 2004.
[4] Daugman, J. G., “Statistical Richness of Visual Phase Information: Update on Recognizing Persons by Iris Patterns,” International Journal of Computer Vision, Vol. 45, No. 1, pp. 25-38, 2001.
[5] Daugman, J. G., “Demodulation by Complex-Valued Wavelets for Stochastic Pattern Recognition,” Int. J. Wavelets, Multi-Resolution Information Processing, Vol. 1, No. 1, pp. 1-17, 2003.
[6] Boles, W. and Boashash, B., “A Human Identification Technique Using Images of The Iris and Wavelet Transform,” IEEE Transactions on Signal Processing, Vol. 46, pp. 1185-1188, Apr., 1998.
[7] Sanchez-Avila, C. and Sanchez-Reillo, R., “Iris-Based Biometrics Recognition Using Dyadic Wavelet Transform,” IEEE Aerospace and Electronic Systems Magazine, Vol. 17, pp. 3-6, Oct., 2002.
[8] Wildes, R., Asmuth, J., Green, G., Hsu, S., Kolczynski, R., Matey, J., and McBride, S., “A Machine-Vision System for Iris Recognition,” Machine Vision and Applications, Vol. 9, No.1, pp. 1-8, 1996.
[9] Zhu, Y., Tan, T., and Wang, Y., “Biometric Personal Identification Based on Iris Patterns,” Proceedings of the International Conference on Pattern Recognition, Barcelona, Spain, Vol. II, pp. 801-804, 2000.
[10]Lim, S., Lee, K., Byeon, O., and Kim, T., “Efficient Iris Recognition Through Improvement of Feature Vector and Classifier,” ETRI Journal, Vol. 23, No. 2, pp. 1-70, 2001.
[11]Ma, L., Wang, Y., and Tan, T., “Iris Recognition Based on Multichannel Gabor Filtering,” Proceedings of the 5th Asian Conference on Computer Vision, Melbourne, Australia, Vol. 1, pp. 279-283, 2002.
[12]Ma, L., Wang, Y., and Tan, T., “Iris Recognition Using Circular Symmetric Filter,” Proceedings of the 16th International Conference on Pattern Recognition, Quebec, Canada, Vol. 2, pp. 414-417, 2002.
[13]Park, C., Lee, J., Smith, M., and Park, K., “Iris-Based Personal Authentication Using a Normalized Directional Energy Feature,” Proceedings of the 4th International Conference, Audio- and Video-Based Biometric Person Authentication, Guildford, UK, Vol. 2688, pp. 224-232, 2003.
[14]Ma, L., Tan, T., Wang, Y., and Zhang, D., “Local Intensity Variation Analysis for Iris Recognition,” Pattern Recognition, Vol. 37, pp. 1287-1298, 2004.
[15]Bae, K., Noh, S., and Kim, J., “Iris Feature Extraction Using Independent Component Analysis,” Proceedings of the 4th International Conference, Audio-and Video-Based Biometric Person Authentication, Guildford, UK, pp. 838-844, 2003.
[16]Ma, Li, Tan, Tieniu, Wang, Yunhong, and Zhang, Dexin, “Personal Identification Based on Iris Texture Analysis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, No. 12, pp. 1519-1533, Dec., 2003.
[ 7]Bracewell, R. N., The Fourier Transform and Its Application, The McGraw-Hill Companies, London, 2000.
[ 8]Casasent, D. and Psaltis, D., “Position, Rotation, and Scale Invariant Optical Correlation,” Applied Optics, Vol. 15, No. 7, pp. 1795-1799, 1976.
[ 9]Mallat, S., A Wavelet Tour of Signal Processing, Academic Press Publishing Company, California, 1998.
[20]Grossman, A. and Morlet, J., “Decomposition of Hardy Functions into Square Integrable Wavelets of Constant Shape,” SIAM Journal of Math. Anal., Vol. 15, No. 4, pp. 723-736, 1984.
[2 ]Fukunaga, K., Introduction to Statistical Pattern Recognition, Academic Press Publishing Company, San Diego, 1990.
[22]Duda, R. O., Hart, P. E., and Stork, D. G., Pattern Classification, Second Edition, Wiley, Canada, 2001.
[23]http://gim.unmc.edu/dxtests/roc2.htm.
[24]Bernier, T. and Jacques-André, L., “A New Method for Representing and Matching Shapes of Natural Objects,” Pattern Recognition, Vol. 36, No. 8, pp. 1711-1723, 2003.
[25]Tsuyoshi, K. and Mohamed, R., “Iris Detection Using Intensity and Edge Information,” Pattern Recognition, Vol. 36, No.2, pp. 549-562, 2003.
[26]Bui, T. D. and Chen, G., “Invariant Fourier-Wavelet Descriptor for Pattern Recognition,” Pattern Recognition, Vol. 32, No. 7, pp. 1083-1088, 1999.
[27]Tang, Y. Y., Li, B. F., Ma, H., and Liu, J., “Ring-Projection-Wavelet-Fractal Signatures, A Novel Approach to Feature Extraction,” IEEE Transactions on Circuits and Systems-II: Analog and Digital Signal Processing, Vol. 45, No. 8, 1998.
[28]Serpico, S. B., “Image and Signal Processing for Remote Sensing VIII,” Proceedings of SPIE, Crete, Greece, Vol. 4855, pp. 285-296, March, 2003.
[29]Institute of Automation, Chinese Academy of Science, CASIA Iris Image Database. http//www.sinobiometrics.com/chinese/chinese.htm.
[30]http://www.cl.cam.ac.uk/users/jgd1000/cameras.html.