臺灣博碩士論文加值系統

在這篇論文中，我們結合了表徵模型和馬可夫模型而提出了一個身份認證的新方法。在我們的身份認證系統中，大致上可以分成兩個主要部份。由於表徵模型不僅可以抽取出人臉上紋理的資訊，並也可以同時抽取出形狀的資訊。因此在第一個部份，我們使用一個表徵模型來抽取出人臉的特徵點。對於身份認證，我們認為人臉形狀的資訊是很有用的。因此，我們先利用一組訓練用的視訊資料來訓練出一個表徵模型，並利用這個表徵模型來抽取出所有人臉影像上的特徵點。為了要建立一個身份認證的系統，我們先使用向量量化的方法來對所有的特徵點做分類，並進一步結合馬可夫模型來充分利用視訊在時間上的資訊。在求出馬可夫模型裡所有的參數之後，為了作身份的認證，我們的方法會動態地去決定出臨界值來判斷是否是為正確的身份。由於身份認證的正確率會被馬可夫模型裡的狀態個數和特徵點的分類個數所影響，因此我們也提出了一個遞迴演算法來適當地決定這兩個變數。
由我們的實驗結果可以發現，我們所提出身份認證系統在我們所自訂的資料庫裡可以有很好的正確率。而在這人臉視訊資庫裡，有六十四個用來訓練系統的視訊資料和六十四個用來測式系統的視訊資料。

In this thesis, we present a novel face authentication scheme by using appearance models and Hidden Markov Models. In our face authentication system, it can be roughly divided into two parts. First, the appearance model is used for features extraction, because an appearance model can not only extract the texture information, but also extract the shape information. We consider the shape information of a face is useful for the face authentication. Thus, we train an appearance model with a training set of labeled image sequences and then use this model to extract the low dimensional features of every image. In order to construct a face authentication system, we apply a vector quantization scheme to classify these features and combine the HMM to make full use of the temporal information across the video sequences. After all parameters in HMM are calculated, we can determine the thresholds dynamically for face authentication. An iterative algorithm with these thresholds is also proposed to select a suitable state number in HMM and a suitable class number of observations, because the performance of face authentication is affected by both variables.
As the result of experiment, we can show that our proposed video-based face authentication system works well on our constructed database. This database contains sixty-four video face sequences for training and sixty-four video face sequences for testing.

Chapter 1: Introduction 1
1.1 Motivation 1
1.2 Organization of This Thesis 3
Chapter 2: Background and Related Work 4
2.1 Face Recognition Using Principal Component Analysis (PCA) 4
2.2 Face Recognition Using Linear Discriminate Analysis (LDA) 7
2.3 Hidden Markov Model (HMM) 10
2.4 Summary 10
Chapter 3: Proposed Face Authentication Approach 15
3.1 System Overview 15
3.2 Introduction to the Appearance Models 15
3.3 Statistical Appearance Model 16
3.3.1 Train the Statistical Models of Shape and Texture 16
3.3.2 Implement the Combined Appearance Models 21
3.3.3 An Example for Facial Appearance Model 22
3.3.4 Given a New Example to Synthesize 24
3.4 Active Shape Model (ASM) 26
3.4.1 Motiviation 26
3.4.2 Model a Local Structure by Using a Set of Training Images 27
3.4.3 Search the Correct Shape of a Face 29
3.5 Active Appearance Model (AAM) 31
3.5.1 Motiviation 31
3.5.2 Train the Prior Information Needed for AAM 32
3.5.3 Search an Example by Using AAM with the Trained Information 35
Chapter 4: Video-based Face Authentication Appraoch 39
4.1 Introduction 39
4.2 Video-Based Face authentication System 41
4.2.1 System Overview 41
4.2.2 Determine the suitable paramters in HMM 42
4.3 Experimental Result 46
Chapter 5: Conclusion and Future Work 51

Reference 52

[1]A. J. GOLDSTEIN, L. D. HARMON and A. B. LESK, “Identification of Human Faces,” Proceedings of the IEEE, vol. 59, no. 5, pp. 748-760, May 1971.
[2]R. CHELLAPPA, C. L. WILSON and S. SIROHEY, “Human and Machine Recognition of Faces: A Survey,” Proceedings of the IEEE, vol. 83, no. 5, pp. 705-741, May 1995.
[3]A. Nefian, “A Hidden Markov Model-Based Approach for Face Detection and Recognition,” PhD thesis, Georgia Institute of Technology, Atlanta, GA, August 1999.
[4]W. A. Barrett, “A Survey of Face Recognition Algorithms and Testing Results,” in Proc. IEEE Int. Conf. Signals, Systems & Computers, vol. 1, pp. 301-305, Nov. 1997.
[5]X. Tang and Z. Li, “Video Based Face Recognition Using Multiple Classifiers,” in Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, pp. 345-349, May 2004.
[6]M. A. Mohamed and P. Gader, “Generalized Hidden Markov Models—Part I: Theoretical Frameworks,” IEEE Trans. Fuzzy Systems, vol. 8, no. 1, pp. 67-81, Feb. 2000.
[7]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.
[8]L. R. RABINER, “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,” Proceedings of the IEEE, vol. 77, no. 2, pp. 257-286, Feb. 1989.
[9]S. Werner and G. Rigoll, “Pseudo 2-dimensional Hidden Markov Models in Speech Recognition,” in Proc. IEEE Int. Conf. Automatic Speech Recognition and Understanding, pp. 441-444, Dec. 2001.
[10]X. Zhou, X. Huang, B. Xu and Y. Wang, “Real-time Facial Expression Recognition Based on Boosted Embedded Hidden Markov Model,” in Proc. IEEE Int. Conf. Image and Graphics, pp. 290-293, Dec. 2004.
[11]H. GUO, Z.-Y. OU and Y. HE, “Automatic fingerprint classification based on embedded Hidden Markov Models,” in Proc. IEEE Int. Conf. Machine Learning and Cybernetics, vol. 5, pp. 3033-3038, Nov. 2003.
[12]A. Kale, A. N. Rajagopalan, N. Cuntoor and V. Kruger, “Gait-based Recognition of Humans Using Continuous HMMs,” in Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, pp. 321-326, May 2002.
[13]X. Liu and T. Chen, “Video-Based Face Recognition Using Adaptive Hidden Markov Models,” in Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, vol. 1, pp. 340-345, June 2003.
[14]T. F. Cootes and C. J. Taylor, “Statistical Models of Appearance for Computer Vision1,” PhD thesis, Manchester, U.K., March 2004.
[15]T. F. Cootes, G. J. Edwards and C. J. Taylor, "A Comparative Evaluation of Active Appearance Model Algorithms," in Proc. British Machine Vision Conference, 1998.
[16]T. F. Cootes, A. Hill, C. J. Taylor and J. Haslam, "The Use of Active Shape Models for Locating Structures in Medical Images," in Image & Vision Computing, July 1994.
[17]H. H. Thodberg and A. Rosholm, “Application of the Active Shape Model in a commercial medical for bone densitometry,” in Proc. BMVC, vol. 1, pp. 43-52, Sep. 2001.
[18]H. Kang, T. F. Cootes and C. J. Taylor, “A Comparison of Face Verification Algorithms using Appearance Models,” in Proc. BMVC, vol. 2, pp. 477-486, 2002.
[19]Z. Li, H. Ai and G. Xu, "Training Support Vector Machines for Video Based Face Recognition," In Proc. IEEE Int. Conf. Image and Graphics, 2002.
[20]Y. Li, S. Gong and H. Liddell, “Video-Based Online Face Recognition Using Identity Surfaces,” in Proc. IEEE Int. Conf. Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, pp. 40-46, July 2001.
[21]K.-C. Lee, J. Ho, M.-H. Yang and D. Kriegman, “Video-Based Face Recognition Using Probabilistic Appearance Manifolds,” in Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, vol. 1, pp. 313-320, June 2003.
[22]X. Tang and Z. Li, “Frame Synchronization and Multi-Level Subspace Analysis for Video Based Face Recognition,” in Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, vol. 2, pp. 902-907, July 2004.