臺灣博碩士論文加值系統

以非侵入式的技術來抽取出測試者的特徵所完成的即時人像辨識系統已經具有極高的辨識正確率。雖然即時人像辨識系統具有許多的優點，然而在一不受限制的辨識環境下進行人像辨識卻是一件極為困難的工作。許多的人像辨識系統必須對影像設下嚴苛的條件如拍攝環境的光線投射角度、影像格式大小、雜訊、以及臉部表情等才能正常運作。本篇論文提出一套使用直線取樣法結合Gabor filters所完成的即時人像辨識系統以降低系統對上述條件的限制。而此演算法對於測試者頭部的偏轉以及影像格式大小改變等具有一定程度的抵抗力。此外，計算的過程也十分有效率。在抽取特徵向量之前，我們先使用Gabor filters將臉部的重要特徵如眼睛、鼻子、嘴唇等予以強化，使其不易受影像採光所影響，接著再使用隨機直線取樣的方式來抓取臉部影像的特徵。最後，使用最鄰近分類器將測試影像正確分類為資料庫中的某一成員。在我們的實驗中，使用ORL人像資料庫來測試此人像辨識系統，測試的結果顯示其平均辨識正確率為99.6 %，平均辨識一張測試影像所需時間為1.784秒。

Real-time face recognition systems based on inoffensive feature extraction techniques have already produced very high identification rates. Although real-time face recognition system has many advantages, real-time face recognition in an unconstrained environment is a difficult task. Many real-time human face recognition systems operate under strict imaging conditions such as controlled illumination, image size, noises, and limited facial expressions. In this thesis, we propose a line-based face recognition algorithm combined with Gabor filters to alleviate the constraints. This algorithm achieves high recognition rates for rotations both in and out of the plane, is robust to sacling, and is computationally efficient. Before the feature vector extraction, we use Gabor filters to enhance the important facial features such as the eyes, the noise, and the mouth. After that, we use a set of random straight lines to extract the feature vector of the face image. Finally, the nearest-neighbor classifier is used to classify the test face image into one of the persons in the database. In our experiment, we use ORL face database to test this line-based face recognition system. Our method achieved an average recognition rate of 99.6 % using 1.784 seconds per view in average.

目錄
第一章緒論 1
1.1 前言………………………………………………………………1
1.2 簡介………………………………………………………………2
1.3 影像處理技術簡介………………………………………………8
1.4 論文大綱 ………………………………………………………10
第二章 文獻回顧與探討 11
2.1 簡介 ……………………………………………………………11
2.2 人像辨識系統 …………………………………………………12
2.3 Gabor filters…………………………………………………16
2.4 聚類演算法 …………………………………………………19
2.4.1 最鄰近演算法…………………………………………20
2.4.2 K最鄰近演算法....………………………………………20
2.4.3 K平均演算法……………………………………………21
第三章 隨機直線取樣法於人像辨識的應用 23
3.1 前言 ……………………………………………………………23
3.2 以隨機直線取樣為基礎的人像辨識演算法………………………24
3.2.1 簡介 …………………………………………………24
3.2.2 人像偵測 ………………………………………………25
3.2.3 訓練階段的隨機直線取樣 ……………………………26
3.2.4 建立訓練資料庫 ………………………………………28
3.2.5 距離 ……………………………………………………29
3.2.6 測試階段的人像偵測 …………………………………30
3.2.7 建立測試資料庫 ………………………………………30
3.2.8 人像辨識 ………………………………………………31
3.3 Gabor filters……………………………………………………35
3.3.1 簡介………………………………………………………35
3.3.2 數學理論…………………………………………………35
3.3.3 Gabor filters的特性…...………………………………39
3.4 即時人像辨識系統 ………………………………………………41
3.4.1 簡介...……………………………………………………41
3.4.2人像偵測...………………………………………………42
3.4.3建立訓練階段的取樣模型 ……………………………43
3.4.4訓練階段的特徵強化 …………………………………44
3.4.5訓練階段的隨機直線取樣 ……………………………45
3.4.6建立訓練資料庫 ………………………………………46
3.4.7測試階段的人像偵測 …………………………………47
3.4.8建立測試階段的取樣模型 ……………………………47
3.4.9建立測試資料庫 ………………………………………48
3.4.10人像辨識 ………………………………………………49
第四章 實驗結果與討論 52
4.1 前言………………………………………………………………52
4.2 人像資料庫………………………………………………………53
4.2.1 ORL人像資料庫………………………………………53
4.2.2 自製小型資料庫 ………………………………………53
4.3 實驗結果與探討……………………………………………………55
4.3.1 第一次修正的直線取樣演算法.…………………………56
4.3.2 隨機直線取樣法結合Gabor filters …………………….58
第五章 結論與未來方向 66
5.1 結論 ………………………………………………………………66
5.2 未來發展 …………………………………………………………67
參考文獻(Bibliography) 68圖目錄
圖 3-1 原始灰階影像 (取自ORL人像資料庫)…………………25
圖 3-2 影像中僅含有人像的部分(i.e. )………………………25
圖 3-3 在影像的人臉部分進行隨機直線取樣………………………26
圖3-4 原始取樣線條經過重新取樣後所得到的值…………………27
圖 3-5 全部共K人的訓練資料庫示意圖…………………………28
圖 3-6 全部共N條取樣線的測試資料庫示意圖…………………30
圖 3-7 文獻[10]的隨機直線取樣演算法流程圖……………………34
圖 3-8 Gabor filters 在頻域中的模型 (取自文獻[11])……..……38
圖 3-9 Gabor filters在頻域中的俯視圖….………………………38
圖 3-10 Gabor filters在頻域中的剖面圖…..………………………39
圖 3-11 在時域中的原始影像...………………………………………40
圖 3-12 經DFT轉換至頻域後所得之影……………………………40
圖 3-13 原始灰階影像 (影像格式為160*160)………………………42
圖 3-14 影像中僅含有人像的部分(i.e. )….…..…………………42
圖3-15 在影像的人臉部分進行直線隨機取樣………………………43
圖 3-16 原始影像與18個Gabor filters 進行迴旋積分…..………44
圖 3-17 在18層經過迴旋積分後的影像上以直線隨機取樣………45
圖 3-18 全部共K人的訓練資料庫示意圖….………………………46
圖 3-19 全部共18層的測試資料庫示意圖…………………………48
圖 3-20 修正後的隨機直線取樣演算法流程圖..……………………..51
圖 4-1 ORL資料庫中某位成員(s1)的影像 ………………………54
圖4-2 臉部大小及頭部偏轉角度校正前後比較圖...………………54
圖 4-3 小型資料庫中某位成員( Roger Yang )的影像……………55
圖 4-4 改變光線後的ORL資料庫…………………………………..58
表目錄
表4-1 第一次修正的隨機取樣演算法實驗結果……………………57
表4-2 第一次修正的隨機取樣演算法實驗結果 (改變光線投射角度 的ORL資料庫)………………………………………………59
表4-3 第二次修正的隨機取樣演算法實驗結果 (改變光線投射角度的ORL資料庫)………………………………………………61
表4-4 經過兩次修正後的演算法應用在原始的ORL資料庫……..63
表4-5 第一次修正後的演算法應用自製的小型資料庫中…………64
表4-6 第二次修正後的演算法應用自製的小型資料庫中…………65

參考文獻
[1] J.G. Daugman, “High confidence visual recognition of persons by a test of statistical independence ,” IEEE Trans. Pattern Anal. Machine Intell., vol.15, pp. 1148-1161, Nov.1993.
[2] R. R. Rogers, “Biometric identification systems：The science of transaction facilitation ,” in Automatic Systems for the Identification and Inspection of Humans, R. J. Mammone and J. D. Murley, Eds., vol.2277 SPIE, 1994, pp.194-199.
[3] 繆紹綱 編著，“數位影像處理 活用Matlab，”全華科技圖書股份有限公司 出版。
[4] G. Cottrell and M. Fleming, “Face recognition using unsupervised feature extraction ,” in Proc. Int. Neural Network Conf.,1990.
[5] J. Buhmann, J. Lange, and C. von der Malsburg, “Distortion invariant object recognition by matching hierarchically labeled graphs, ” in Proc. IJCNN’89, 1989, pp.151-159.
[6] A.L. Yuille, “Deformable templates for face recognition,” J. Cognitive Neurosci., vol.3, no.1, pp.59-70, 1991.
[7] M. Turk and A. Pentland, “Eigenfaces for recognition,” J. Cognitive Neurosci., vol.3, no.1, pp.71-86, 1991.
[8] Z.-Q. Hong, “Algebraic feature extraction of image for recognition,” Patt. Recogn., vol.24, no.3, pp.211-219, 1991.
[9] O. Nakamura, S. Mathur, and T. Minami, “Identification of human faces based on isodensity maps,” Patt. Recogn., vol.24, no.3, pp.263-272, 1991.
[10] O. De Vel, and S. Aeberhard,“Line-Based Face Recognition under Varying Pose,”IEEE Trans. Pattern Analysis and Machine Intelligence, vol.21, no.10, pp.1081—1088 , Oct. 1999.
[11] B. Duc, S. Fischer,and J. Bigün,“Face Authentication with Gabor Information on Deformable Graphs,”IEEE Trans. Image Processing, vol.8, no.4, pp. 504-516, April 1999.
[12] 林宸生 編著，“數位信號…影像與影音處理”，高立書局出版，1998。
[13] R. Brunelli ,“Estimation of Pose and Illumination Direction for Face Processing,”Image and Vision Computing , vol.15, pp.741-748, 1997.
[14] T. Kanade , “ Picture Processing by Computer Complex and Recognition of Human Faces,” Technical report , Dept of Information Sciences , Kyoto Univ.,1973.
[15] R. Brunelli and T. Poggio ,“Face recognition：Features versus Templates,”IEEE Trans. Pattern Analysis and Machine Intelligence, vol.15, no.10, pp.1042—1052, Oct. 1993.
[16] M. Turk , and A.Pentland , “Eigenfaces for Recognition,” Jounral of Cognitive Neuro-science , 3(1):71-86 , 1991.
[17] P.N. Belhumeur, J.P. Hespanha, and D.J. Kriegman ,“Eigenfaces vs. Fisherfaces：Recognition using Class Specific Linear Projection,”IEEE Trans. Pattern Analysis and Machine Intelligence, vol.19, no.7, pp.711—720, July 1997.
[18] K.M. Lam, H. Yan ,“An Analytic-to-Holistic Approach for Face Recognition Based on A Single Frontal View,”IEEE Trans. Pattern Analysis and Machine Intelligence, vol.20, no.7, pp.673—686, July 1998.
[19] V.V. Kohir, U.B. Desai,“Face Recognition using a DCT-HMM Approach，”Fourth IEEE Workshop on,Applications of Computer Vision,1998. WACV '98. Proceedings., pp. 226 —231, 1998.
[20] S. Edelman, D. Reisfield, and Y. Yeshurun, “ Learning to Recognize Faces from Examples,” Lecture Notes in Computer Science, vol.588, pp.787-791, 1992.
[21] M. Zhang, and J. Fulcher,“ Face Recognition using Artificial Neural Network Group-Based Adaptive Tolerance (GAT) Trees ,” IEEE Trans. Neural Networks, vol.7, no.3, pp. 555—567, May 1996.
[22] A.R. Mirhosseini, H. Yau, C. Chen, and T. Pham,“Human Face Recognition：A Minimal Evidence Approach”,Sixth International Conference on, Computer Vision, pp.652—659, 1998.
[23] Y. Yacoob, and L.S. Davis,“Recognizing Human Facial Expressions from Long Image Sequences using Optical Flow,” IEEE Trans. Pattern Analysis and Machine Intelligence , vol.18, no.6, pp.636 —642, June 1996.
[24] B. Achermann and H. Bunke, “Combination of Face Classifiers for Person Identification, ” Technical Report IAM-96-002, Institut für Informatik und Angewandte Mathematik, Universität Bern, 1996.
[25] J. Zhang, Y. Yan, and M.Lades,“Face Recognition：Eigenface, Elastic Matching, and Neural Nets,”Proc. IEEE, vol.85, no.9, pp.1423—1435, Sept. 1997.
[26] D.Gabor, “Theory of Communications ,” J. Inst. Electr. Eng., vol.93, pp.429-457, 1946.
[27] S. Marcelja, “Mathematical Description of the Responses of Simple Cortical Cells,” Journal of the Optical Society of America, vol. 70, no.11, pp.1297-1300, 1980.
[28] J. G. Daugman, “Uncertainty Relation for Resolution in Space, Spatial Frequency, and Orientation Optimized by Two-Dimensional Visual Cortical Filters,” Journal of the Optical Society of America, vol.2, no.7, pp.1160-1169, 1985.
[29] R. Mehrotra, K.R. Namuduri and N. Ranganathan, “Gabor Filter-Based Edge Detection ,” Pattern Recognition, vol.25, no. 12, pp.1479-1494, 1992.
[30] R. Buse, and Z.-Q. Liu,“Feature Measurement And Analysis using Gabor Filters,”1995 International Conference on, Acoustics, Speech, and Signal Processing, 1995. ICASSP-95. , vol .4 , pp. 2447-2450, 1995.
[31] K.R. Namuduri, R. Mehrotra, and N. Ranganathan,“Edge Detection Models Based on Gabor Filters,”International Conference on , Pattern Recognition Vol.III. Conference C: Image, Speech and Signal Analysis, Proceedings., 11th IAPR 1992 , pp.729-732, 1992.
[32] J.F. Canny, A Computational Approach to Edge Detection, IEEE Trans. on PAMI, vol.8, no.6, pp.679-698, 1986.
[33] C.-J. Lee and S.-D. Wang,“A Gabor Filter-Based Approach to Fingerprint Recognition,”1999 IEEE Workshop on. Signal Processing Systems,1999. SiPS 99. pp.371 —378 , 1999.
[34] K.-C. Chung, S. C. Kee, and S. R. Kim,“Face Recognition using Principal Component Analysis of Gabor Filter Responses,” International Workshop on ,Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, 1999. Proceedings. pp.53 —57, 1999 .
[35] J. Mailk and P. Perona, “Preattentive texture discrimination with early vision mechanisms,” J.Opt.Soc.Amer., vol.7, pp.923-932, May 1990.
[36] J. Bigün and J. M. H. du Buf, “N-folded symmetries by complex moments in Gabor space,” IEEE Trans. Pattern Anal. Machine Intell., vol.16, pp.80-87, Jan. 1994.
[37] 林鈺山 編著，“模糊類神經網路在人像偵測與辨識之應用，” 中興大學電機工程研究所碩士論文，1999。