近來影像技術因電腦運算速度的提昇，其運用也越來越廣泛，尤其以視訊會議系統、人臉辨識、電腦視覺等多有研究，然而以往的相關研究中多以靜止的攝影機、單純的背景來做追蹤，本論文以倒傳遞類神經網路搭配影像處理技術，以及自製動態攝影平台作臉部追蹤與辨識的研究。在人臉追蹤方面，利用倒傳遞類神經網路來將膚色分割，將膚色從複雜的背景中分離出來，再配合橢圓遮罩搜尋法，找出人臉的所在區域並加運算及追蹤。辨識方面使用主要特徵值與特徵向量為辨識的依據，將畫面中人臉取出與資料庫進行比對，利用最小平方誤差辨識出畫面最可能為何人。經實驗結果證實，本研究提出之人臉追蹤與辨識系統可以有效地尋找出畫面中的人臉區域並標定其中心位置追蹤並正確地辨識目標物。

Face tracking and pose estimation, is getting more and more important recently because it can be applied in a variety of applications, such as the face recognition, video coding and teleconference, etc. However, over the past, the study focused on face tracking, which mostly used the fixed camera with pure background. Inspired by the lack of developments in face tracking, an efficient face tracking and recognition system based on the back propagation neural network of skin color segmentation is proposed in this study.
In face tracking, the back propagation neural network on skin color segmentation is employed to separate the skin color area from a complicated background and the face position is located by elliptic masking. In face recognition, the main eigenvalue and eigenvector of image are initially saved in database as the recognize criterion, and then we use the mean square error to recognize who is in the image. Practically, this system can maintain the face on the center of the screen rapidly as well as provide efficiency for seeking and finding the accurate face and recognition in image.

[1] A. Pentland, B. Moghaddam, and T. Starner, "View-Based and Modular Eigenspaces for Face Recognition", Technical report 245, MIT Media Lab Vismod, 1993.
[2] C.Kervrann, F.Davoine, P.Perez, R.Forchheimer, C.Labit ,“Generalized likelihood ratio-based face detectionand extraction of mouth features”, Pattern Recognition Letters 18, 1997 , 899-912
[3] C.H. Lin; J.L. Wu, ”Automatic facial feature extraction by genetic algorithms”, Image Processing, IEEE Transactions on Volume: 8 6 , June 1999 , Page(s): 834 —845
[4] Gary R. Bradski, "Computer Vision Face Tracking For Use in a Perceptual User Interface", Intel Technology Journal Q2 '98
[5] G.D. Hager, P.N. Belhumeur, "Real-time tracking of image regions with changes in geometry andillumination ", Computer Vision and Pattern Recognition, 1996. Proceedings CVPR '96, 1996IEEE Computer Society Conference on , 1996 , Page(s): 403 -410
[6] Hideaki Ishii, Minour Fukumi, Norio Akamatsu, "Face detection based on skin color information in visual scenes by neural networks", 0-7803-5731-0, 1999 IEEE
[7] J. Luettin, N.A. Thacker, S.W. Beet, "Locating and tracking facial speech features”,Pattern Recognition, 996., Proceedings of the 13th International Conference on Volume: 1, Page(s): 652 -656 vol.1, 1996
[8] J. Sullivan, A. Blake, M. Isard and J. MaCormick, "Object Localization by Bayesian Correlation",Computer Vision, 1999. The Proceedings of the Seventh IEEE International Conference on Volume: 2 , 1999 , Page(s): 1068 -1075 vol.2
[9] J. Heinzmann, Zelinsky," A. 3-D Facial Pose and Gaze Point Estimation using a Robust Real-Time Tracking Paradigm ",Automatic Face and Gesture Recognition, 1998. Proceedings. Third IEEE International Conference on , 1998 , Page(s): 142 —147
[10] J. Luettin, N.A. Thacker, S.W. Beet, "Locating and tracking facial speech features”,Pattern Recognition, 996., Proceedings of the 13th International Conference on Volume: 1, Page(s): 652 -656 vol.1, 1996
[11] J. Luettin, A. Thacker, W.B. Steve, "Locating and Tracking Facial Speech Features",Proceedings of the International Conference on Pattern Recognition, Vienna, Austria, 1996
[12] J. Sullivan, A. Blake, M. Isard and J. MaCormick, "Object Localization by Bayesian Correlation",Computer Vision, 1999. The Proceedings of the Seventh IEEE International Conference on Volume: 2 , 1999 , Page(s): 1068 -1075 vol.2
[13] K. Choong Yow, R. Cipolla, "Feature-Based Human Face Detection", Department of Enginneering,Cambridge
[14] K. Toyama, G.Haga. "Incremental focus of attention for robust visual tracking.", IN Computer Vision and Patt. Recog., pages 189-195,1996
[15] L. Shihong; Y. Sumi, M. Kawade, F. Tomita, " Building 3D facial models and detecting face pose in 3D space ", 3-D Digital Imaging and Modeling, 1999. Proceedings. Second International Conference on , 1999, Page(s): 398 -404
[16] L. Sirovich and M. Kirby, "Low-dimensional procedure for the characterization of human faces", J. Opt. Soc. Am. A / vol.4, No.3 / March 1987.
[17] M. La Cassia, J. Isidoro, S. Sclaroff." Head tracking via robust registration in texture map images", CVPR, 1998.
[18] M. La Cassia, S. Sclaroff. " Fast, reliable head tracking under varying illumination Computer Vision andPattern Recognition", 1999. IEEE Computer Society Conference on. Volume: 1 , 1999 , Page(s): 604 —6
[19] M. Isard, A. Blake. "The Condensation Algorithm", http://www.robots.ox.ac.uk
[20] M. La Cassia, J. Isidoro, S. Sclaroff." Head tracking via robust registration in texture map images", CVPR, 1998.
[21] P.A. Beardsley, “A qualitative approach to classifying gaze direction”, Automatic Face and GestureRecognition, 1998. Proceedings. Third IEEE International Conference on , 1998 , Page(s): 160 —165
[22] P.A. Beardsley, “A qualitative approach to classifying head and eye pose”, Applications of Computer Vision, 1998. WACV '98. Proceedings., Fourth IEEE Workshop on , 1998 , Page(s): 208 —213
[23] P.A. Beardsley, “A qualitative approach to classifying head and eye pose”, Applications of Computer Vision, 1998. WACV '98. Proceedings., Fourth IEEE Workshop on , 1998 , Page(s): 208 —213
[24] R. Kaucic, A. Blake, "Accurate, real-time, unadorned lip tracking", Computer Vision, 1998.Sixth International Conference on , 1998 , Page(s): 370 -375
[25] R. Kaucic, D. Reynard, A. Blake, "Real-time lip trackers for use in audio-visual speechrecognition ", Integrated Audio-Visual Processing for Recognition, Synthesis andCommunication (Digest No: 1996/213), IEE Colloquium on , 1996 , Page(s): 3/1 -3/6
[26] Rama Chellappa, Charles L. Wilson, and Saad Sirohey , "Human and Machine Recognition of Faces: A Survey", Proceeding of the IEEE, Vol. 83, No.5, pp. 705-740, MAY 1995.
[27] Stephen Karungaru, Minour Fukumi and Norio akamatsu, "Detection of Human Faces in Visual Scenes", seven Australian and New Zealand Intelligent Information System Conference, 18-21 November 2001, Perth, Westem Australia
[28] S. Clippingdale, T. Ito," A unified approach to video face detection, tracking and ecognition ", Image Processing, 1999. ICIP 99. Proceedings. 1999 International Conference on Volume: 1 , 1999 , Page(s): 662 -666 vol.1
[29] Shang-Hung Lin and S.Y. Kung, "Probabilistic DBNN via expectation maximization with multi-sensor classification applications", ICIP'95.
[30] 王建權 "以數位影像處理搭配動態攝影機做人臉自動追蹤" 國立成功大學電機工程系碩士論文
[31] 連國珍 "數位影像處裡" 儒林圖書 2002
[32] 黃文吉 "C++ Builder與影像處理" 儒林圖書2005
[33] 黃國峰,張真誠,陳同孝 "數位影像處理技術" 旗標圖書 2004
[34] 葉怡成 "類神經網路模式應用與實作" 儒林圖書 2003
[35] 譚永恆 "以數位影像處理技術作人臉自動追蹤" 國立成功大學電機工程系碩士論文
[36] 繆紹剛 "數位影像處理 活用MATLAB" 全華圖書 2005
[37] 鐘國亮 "影像處理與電腦視覺" 東華書局 2006
[38] 數學傳播,20卷2期,85年6月