在本篇論文中，我們設計與實作一個基於入侵者偵測之半監督式人臉辨識系統，將人臉辨識系統結合監督式訓練及非監督式的訓練。傳統人臉辨識系統主要將訓練及辨識分為兩個階段進行，屬於監督式的學習方法，也因此造成系統的限制：分類器一旦訓練完成，除了調整資料庫重新訓練外，無法在系統運行時進行學習及訓練，也無法自行調整及增加資料庫的資料。本研究之目的是，當新進資料進入系統的同時，系統收集資料並加以分析。發現新類別時，自動在資料庫增加”新”類別，能夠有效更新資料庫，並且減少系統管理人員整理資料的時間，進而提升系統的安全性及彈性。實驗部分顯示我們提出的半監督式人臉辨識系統，能自動調整資料庫及自動學習新的類別，並重新訓練更適合的新分類器。以AT&T資料庫為實驗對象時，當訓練類別為所有類別的80%時，整體辨識率 (包含未訓練的20%類別資料)，可達91.72%，而訓練類別為所有類別的60%時，辨識率 (包含未訓練的40%類別資料) 為81.75%，顯示本研究對新進資料及新類別的學習能力及系統的彈性。

In this thesis, we will present a semi-supervised face recognition system for detecting intruders. In general, the processing of face recognition systems can be divided into two phases: training and recognition. When the classifier training is completed, the system cannot learn new faces and unable to adjust and increase the database. The purpose of this study therefore is when the new information entering, the system will collect and analysis the information. When the system finds a new category, it will automatically update the database. For this reason, the system manager can effectively to label the new images, and it will enhance system’s security and flexibility. The experiments show that our face recognition system has the abilities to update the information of database and re-train a classifier that more suitable. For the benchmark of AT&T, the recognition rate of our proposed system is 91.72% when the training set has 80% categories, and 81.75% of 60% categories, respectively. The results show that our face recognition system is more scalable than traditional face recognition system.

參考文獻
[1] Z. Wen Yi and R. Chellappa, "Illumination-insensitive face recognition using symmetric shape-from-shading," in Computer Vision and Pattern Recognition, 2000. Proceedings. IEEE Conference on, 2000, pp. 286-293 vol.1.
[2] S. Shan, W. Gao, B. Cao, and D. Zhao, "Illumination Normalization for Robust Face Recognition Against Varying Lighting Conditions," in Proceedings of the IEEE International Workshop on Analysis and Modeling of Faces and Gestures: IEEE Computer Society, 2003.
[3] X. Xie and K.-M. Lam, "Face recognition under varying illumination based on a 2D face shape model," Pattern Recognition, vol. 38, pp. 221-230, 2005.
[4] A. M. Bronstein, M. M. Bronstein, and R. Kimmel, "Expression-invariant face recognition via spherical embedding," in Image Processing, 2005. ICIP 2005. IEEE International Conference on, 2005, pp. III-756-9.
[5] A. M. Bronstein, M. M. Bronstein, and R. Kimmel, "Expression-Invariant Representations of Faces," Image Processing, IEEE Transactions on, vol. 16, pp. 188-197, 2007.
[6] B. Gokberk, L. Akarun, and E. Alpaydin, "Feature selection for pose invariant face recognition," in Pattern Recognition, 2002. Proceedings. 16th International Conference on, 2002, pp. 306-309 vol.4.
[7] L. Xiaoming and C. Tsuhan, "Pose-robust face recognition using geometry assisted probabilistic modeling," in Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on, 2005, pp. 502-509 vol. 1.
[8] Y. Wang, L. Wu, L. Tu, and X. Wu, "A face recognition method robust to pose variation," in Signal Processing, 2008. ICSP 2008. 9th International Conference on, 2008, pp. 1600-1603.
[9] W. Zhao, R. Chellappa, P. J. Phillips, and A. Rosenfeld, "Face recognition: A literature survey," ACM Comput. Surv., vol. 35, pp. 399-458, 2003.
[10] C. Cortes and V. Vapnik, "Support-Vector Networks," Mach. Learn., vol. 20, pp. 273-297, 1995.
[11] in AT&T face database, Available: http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html.
[12] M. Turk and A. Pentland, "Eigenfaces for recognition," J. Cognitive Neuroscience, vol. 3, pp. 71-86, 1991.
[13] M. A. Turk and A. P. Pentland, "Face recognition using eigenfaces," in Computer Vision and Pattern Recognition, 1991. Proceedings CVPR '91., IEEE Computer Society Conference on, 1991, pp. 586-591.
[14] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, "Eigenfaces vs. Fisherfaces: recognition using class specific linear projection," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 19, pp. 711-720, 1997.
[15] P. Penev and J. Atick, "Local feature analysis: A general statistical theory for object representation," Network: computation in neural systems, vol. 7, pp. 477-500, 1996.
[16] T. Ahonen, A. Hadid, and M. Pietikainen, "Face Description with Local Binary Patterns: Application to Face Recognition," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 28, pp. 2037-2041, 2006.
[17] T. Cover and P. Hart, "Nearest neighbor pattern classification," Information Theory, IEEE Transactions on, vol. 13, pp. 21-27, 1967.
[18] B. Moghaddam, T. Jebara, and A. Pentland, "Bayesian face recognition," Pattern Recognition, vol. 33, pp. 1771-1782, 2000.
[19] B. Moghaddam, W. Wahid, and A. Pentl, "Beyond eigenfaces: Probabilistic matching for face recognition," 1998.
[20] V. Vapnik, "Statistical learning theory. 1998," NY Wiley.
[21] S. Fidler and A. Leonardis, "Robust LDA Classification by Subsampling," in Computer Vision and Pattern Recognition Workshop, 2003. CVPRW '03. Conference on, 2003, pp. 97-97.
[22] S. Fidler, D. Skocaj, and A. Leonardis, "Combining reconstructive and discriminative subspace methods for robust classification and regression by subsampling," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 28, pp. 337-350, 2006.
[23] W. A. Khan, M. Y. Javed, and M. A. Anjum, "Occluded Face Images Recognition Using Robust LDA," in Emerging Technologies, 2006. ICET '06. International Conference on, 2006, pp. 151-156.
[24] T. Ojala, M. Pietikainen, and D. Harwood, "A comparative study of texture measures with classification based on featured distributions," Pattern Recognition, vol. 29, pp. 51-59, 1996.
[25] T. Bayes, "An essay towards solving a problem in the doctrine of chances. 1763," MD Comput, vol. 8, pp. 157-71, May-Jun 1991.
[26] J. Platt, N. Cristianini, and J. Shawe-Taylor, "Large margin DAGs for multiclass classification," Advances in neural information processing systems, vol. 12, pp. 547-553, 2000.
[27] K. Shivsubramani, R. Loganathan, C. Srinivasan, V. Ajay, and K. Soman, "Multiclass Hierarchical SVM for Recognition of Printed Tamil Characters."
[28] V. Vural and J. Dy, "A hierarchical method for multi-class support vector machines," 2004.
[29] S. Liu, H. Yi, L. Chia, and D. Rajan, "Adaptive hierarchical multi-class SVM classifier for texture-based image classification," 2005, p. 4.
[30] K. Benabdeslem and Y. Bennani, "Dendogram-based SVM for multi-class classification," Journal of Computing and Information Technology, vol. 14, p. 283, 2006.
[31] L. Cheng, J. Zhang, J. Yang, and J. Ma, "An Improved Hierarchical Multi-class Support Vector Machine with Binary Tree Architecture," 2008, pp. 106-109.
[32] B. Liu, "Web data mining," Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data, Data-Centric Systems and Applications, Volume. ISBN 978-3-540-37881-5. Springer Berlin Heidelberg, 2007, 2007.