機器視覺被使用在許多自動化系統上，諸如影像監督、臉部辨識與行為監測。對視覺監控系統而言，目標追蹤器是一種可以用來管理來源影像新進資料的重要技術。而在真實環境的條件下，通常並不會只考慮單一的追蹤目標，所以一個可以追蹤多重目標的追蹤系統是必須的。臉部偵測是常見的機器視覺技術之ㄧ，可被應用在監視系統、辨識系統與機器人視覺上。在這篇論文中，我們提出了一個可追蹤多重人臉且能同時定位人臉五官的系統。我們使用HSV色彩空間來選取膚色區塊，並利用形狀與密度資訊來得到候選的人臉區塊，在擷取臉部五官的步驟之前，人臉五官的拓樸邏輯可先幫我們預測可能的嘴部與眼睛位置。在擷取出五官之後，為了追蹤臉部的偵測結果，我們提出了一個結合了粒子濾除器與多重假設追蹤演算法概念的多重目標追蹤系統。我們使用速率資訊改進了粒子濾除器的追蹤效能，使其可適應快速移動之物體，另外使用多重假設追蹤演算法中的分群概念來區分目標，當一個群中有兩個以上的目標時，使用歷史軌跡紀錄的資訊來做兩物體相交時的追蹤預測。由於視訊畫面的偵測資料必須定時更新以符合即時多重追蹤，當有更新資料進來時，我們使用多重假設追蹤的概念來連結新資料與舊有目標以及更新目標。根據實驗結果，我們的系統可以有效即時追蹤多重人臉與五官。

Machine vision can be used in many automatic systems, like video surveillance, face recognition, and pedestrian monitoring. For a vision monitor system, the target tracker is an important technology to manipulate the coming data from source images. In real environments, there is not usually a single target to be considered, and a tracker for multiple targets is needed. Face detection is one of commonly used machine vision technologies applied to monitoring systems, recognition systems, and robot vision. In this thesis, we propose a multi-face tracking system which can also locate facial features of each face. We use the HSV color space to select skin-color regions, then employ shape and density information to choose candidate face regions. Before extracting the facial features from the candidate regions, the topology human facial features can help us to predict possible positions of a mouth and eyes. After facial features extraction, we propose a multiple targets tracking system combined the conceptions of a particle filter and the multiple hypotheses tracking (MHT) algorithm to track the target face. We use trajectory information to improve the performance of particle filter, and let it can adaptive to fast motion objects. We group the targets by the clustering conception of the MHT algorithm. If there are multiple targets belong to the same cluster, we predict the tracking positions by history trajectory records when two objects crossover. Because the detection data of a video image sequence must be updated to suit the real-time multiple targets tracking, we exploit use the conception of the MHT algorithm to associate the new measurements and prior targets for updating targets information when a new set of data is coming. According to our experimental results, our system can efficiently track multiple faces with facial features in real time.

[1] R. L. Hsu, M. Abdel-Mottaleb, and A. K. Jain, “Face detection in color images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.24, no.5, pp.696-706, 2002.
[2] C. Y. Su, Y. T. Zhuang, L. Huang, and F. Wu, “A Two-Step Approach to Multiple Facial Features Tracking: Temporal Particle Filter and Spatial Belief Propagation,” IEEE International Conference on Automatic Face and Gesture Recognition, 2004.
[3] Marquardt Beauty Analysis, California. http://www.beautyanalysis.com/
[4] D. B. Reid, “An Algorithm for Tracking Multiple Targets,” IEEE Transactions on Automatic Control, vol.AC-24, no.6, pp.843-854, 1979.
[5] R. Danchick, and G. E. Newnam, “A Fast Method for Finding the Exact N-best Hypothesis for Multitarget Tracking,” IEEE Transactions on Aerospace and Electronic Systems vol.29, no.2, pp.555-560, 1993.
[6] K. C. Chang, S. Mori, and C. Y. Chong, “Evaluating a Multiple-Hypothesis Multitarget Tracking Algorithm,” IEEE Transactions on Aerospace and Electronic System, vol.30, no.2, pp.578-590, 1994.
[7] M. Yeasin, E. Polat, and R. Sharma, “A Multiobject Tracking Framework for Interactive Multimedia Applications,” IEEE Transactions on Multimedia, vol.6, no.3, pp.398-405, 2004.
[8] I. J. Cox, and S. L. Hingorani, “An Efficient Implementation of Reid’s Multiple Hypothesis Tracking Algorithm and Its Evaluation for Purpose of Visual Tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.18, no.2, pp.138-150, 1996.
[9] 廖偉全, 以電腦視覺為基礎之眼睛定位與追蹤技術於即時的眨眼偵測, 碩士學位論文, 國立台灣科技大學自動化及控制研究所, 台北市, 2005.
[10] 王凱毅, “A Real-time Face Tracking and Recognition System Based on Particle Filter and AdaBoosting Techniques,” 碩士學位論文, 國立台灣科技大學資訊工程系, 台北市, 2006.
[11] S. Caifeng, W. Yucheng, T. Tieniu, and F. Ojardias, “Real time hand tracking by combining particle filtering and mean shift,” in Proceedings of the Sixth IEEE International Conference on Automatic Face and Gesture Recognition (FGR’04), pp. 669-674, 2004.
[12] M. H. Yang, D. Kriegman, and N. Ahuja, “Detecting faces in images: a survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 1, pp. 34-58, 2002.
[13] G. Yang and T. S. Huang, “Human face detection in complex background,” in Pattern Recognition, vol. 27, no. 1, pp. 53-64, 1994.
[14] K. C. Yow and R. Cipolla, “Feature-based human face detection,” Image and Vision Computing, vol. 15, no. 9, pp. 712-735, 1997.
[15] D. Chai and A. Bouzerdoum, “A Bayesian approach to skin color classification in YcbCr color space,” in Proceedings of the IEEE Region Ten Conference, vol. 2, pp. 421-424, 2000.
[16] A. Lanitis, C. J. Taylor, and T. F. Cootes, “An automatic face identification system using flexible appearance models,” Image and Vision Computing, vol. 13, no. 5, pp. 393-401, 1995.
[17] R. Vaillant, C. Monrocq, and Y. Le Cun “An original approach for the localization of objects in images,” in Proceedings of the IEEE Conference on Artificial Neural Networks, pp. 245-250, 1994.
[18] M. Turk and A. Pentland, “Eigenfaces for recognition,” Journal of Cognitive Neuroscence, vol. 3, no. 1, pp. 71-86, 1991.
[19] S. Young-Bum, K. Jin-Tae, and L. Won-Hyung, “Extraction of face objects using skin color information,” in Proc. of IEEE Int. Conf. on Communications, Circuits and Systems and West Sino Expositions, vol.1, pp.600-604, 2002.
[20] H. G. Kim, N. H. Kim, and S. C. Ahn, “Skin Region Extraction Using Moving Color Technique,” The 4th IEEE International Workshop on ISPACS ’98, Nov. 1998.
[21] E. Loutas, I. Pitas, and C. Nikou, “Probabilistic Multiple Face Detection and Tracking Using Entropy Measures,” in IEEE Transactions on Circuits and System for Video Technology, vol.14, no.1, pp.128-135, 2004.
[22] S. Kawato, and J. Ohya, “Real-time detection of nodding and head-shaking by directly and tracking between-eyes,” Proc. IEEE Fourth Int. Conf. Automatic Face Gesture Recognition, pp.40-45, 2000.
[23] S. Kawato, and N. Tetsutani, “Detection and tracking of eyes for gaze-camera control,” in Image and Vision Computing, vol.22, pp.1031-1038, 2004.
[24] H. Tan, and Y-J Zhang, “Detecting eye blink states by tracking iris and eyelids,” in Pattern Recognition Letters, vol.27, pp.667-675, 2006.
[25] T. Brandt, R. Stemmer, and A. Rakotonirainy, “Affordable Visual Driver Monitoring System for Fatigue and Monotony,” in IEEE International Conference on Systems, Man and Cybernetics, pp.6451-6456, 2004.
[26] 范國清, 王元凱, 陳炳富, 追蹤演算法簡介, 影像與識別, 第8卷, 第4期, 第17-30頁, 2002.
[27] K. H. An, D. H. Yoo, S. U. Jung, and M. J. Chung, “Robust multi-viewface tracking,” in Proceedings of the IEEE International Conference on Intelligent Robots and Systems, pp. 1905-1910, 2005.
[28] S. Birchfield, “Elliptical head tracking using intensity gradients and color histograms,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 232-237, 1998.
[29] P. Fieguth and D. Terzopoulos. “Color-based tracking of heads and other mobile objects at video frame rates,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 21-27, 1997.
[30] R. M. Haralick and L. G. Shapiro, Computer and Robot Vision, vol. 2, Addison-Wesley Reading, Massachusetts, 1993.
[31] C. Y. Tang, Y. P. Hung, and Z. Chen, “Automatic detection and tracking of human head using an active stereo vision system,” in Proceedings of Asian Conference on Computer Vision, vol. 1, pp. 632-639, 1998.
[32] N. Herodotou, K. N. Plataniotis, and A. N. Venetsanopoulos, “Automatic location and tracking of the facial region in color video sequences,” Signal Processing and Image Communication, pp. 359-388, 1999.
[33] M. Montemerlo, S. Thrun, and W. Whittaker, “Conditional particle filters for simultaneous mobile robot localization and people-tracking,” in Proceedings of the IEEE International Conference on Robotics and Automation, vol. 1 , pp. 695-701, 2002.
[34] K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image and Vision Computing, vol. 21, no. 1, pp. 99-110, 2003.
[35] R. L.Popp, K. R. Pattipati, and Y.B-S, “m-Best S-D Assignment Algorithm to Multitarget Tracking,” in IEEE Transactions on Aerospace and Electronic Systems, vol.37, no. 1, pp.22-39,2001.
[36] L. Jing and P. Vadakkepat, “Multiple Targets Tracking by Optimized Particle Filter Based on Multi-scan JPDA,” in Instrumentation and Measurement Technology Conference, pp.303-308, 2004.