在電腦視覺領域中，有關物件分類與人類動作分析一直都是重要的研究課題。其目的是希望透過電腦視覺的方法，在不需要人為之操作下，讓系統能夠自動對攝影機所擷取的影像進行目標物偵測與追蹤，並進一步分析其物件種類或動作行為。當系統偵測出異常行為時可於第一時間發出警告，使其能符合即時安全監控系統之要求。在目標物偵測與追蹤方面，本論文使用一改良式適應性背景相減法取出目標物，並使用一陰影偵測演算法來消除目標物之陰影部份，以獲得更佳的目標物輪廓。此外使用一具有多重相似度量測之追蹤模組在連續畫面中追蹤各個目標物。在物件分類方面，本論文採用模糊平均分類法，計算類別的群聚中心及特徵，並分辨出移動目標物之類型。而在人類動作分析方面，因人體並無固定幾何形狀，故使用人體輪廓投影圖做為特徵，採用機率式姿態分類器進行姿態分析並紀錄在時間序列上，再根據事先建立之姿態狀態轉換及模糊規則進行比對，給予相對應之動作。最後透過數個監控影片進行系統的測試，實驗結果顯示本論文所採用之演算法確實可行。

Moving object classification and human activity analysis have always been import research topics in the field of computer vision. The aim of these two research topics is to exploit the computer vision so that the system can automatically detect, track and classify the moving target, and also analyze its action. When an abnormal event is detected, a warning message will be issued immediately so that the requirement of real-time security and surveillance can be satisfied. As for the problem of object detection and tracking, a modified adaptive background subtraction method is employed to detect the moving object in this study. In addition, a shadow detection algorithm is used to eliminate parts of the shadows in the moving object in order to improve the quality of the detected object contour. Moreover, a tracking module that uses multi-cue similarity matching is used to track the moving object in a sequence of video frames. For moving object classification, the Fuzzy C-Means algorithm is used to calculate the center of the cluster and the feature vector to classify the moving object in this study. As for the problem of human activity analysis, due to the fact that the human body has a non-rigid shape, the study utilizes the projection histograms of the human body to analyze the human postures and recorded in a time sequence. Based on the matching results between the pre-built state-transition graph and the posture of time sequence using fuzzy inference, an appropriate action will be assigned. Experimental results show that the adopted approach exhibits satisfactory performances.

[1] A. Prati, I. Mikic, M.M. Trivedi, and R. Cucchiara, “Detecting moving shadows: algorithms and evaluation,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 25, no. 7, pp. 918–923, 2003.
[2] 王俊凱，基於改良式適應性背景相減法與多重影像特徵比對法之多功能及時視覺追蹤系統之設計與實現，碩士論文，國立成功大學電機工程學系，2004。
[3] R. Cucchiara, C. Grana, A. Prati and R. Vezzani, “Probabilistic Posture Classification for Human Behavior Analysis,” IEEE Trans. on Systems Man and Cybernetics part A- Systems and Humans, vol. 35, no. 1, pp.42- 54, 2005
[4] R. Cucchiara, C. Grana, A. Prati and R. Vezzani, ”Computer vision system for in-house video surveillance, ” IEE Proceedings -Vision Image and Signal Processing, vol. 152, no. 2, pp. 242-249, 2005.
[5] L. Panini, R. Cucchiara, “A machine learning approach for human posture detection in domotics applications, ” in Proc. of the IEEE Conference on Image Analysis and Processing, 2003, pp. 103 - 108.
[6] H. Ushida, A Imura, T Yamaguchi and T. Takagi, “Human-Motion Recognition by means of Fuzzy Associative Inference,” in Proc. of IEEE International Conference on Fuzzy System, 1994, pp. 813, vol. 2.
[7] M-C Su, “A Fuzzy Rule-Based Approach to Spatio-Temporal Hand Gesture Recognition,” IEEE Trans. on Systems Man and Cybernetics part C - Applications and Reviews, vol. 30, no. 2, pp.276-281, 2000.
[8] M. Oren, C. Papageorgiou, P. Sinha, E. Osuna and T. Poggio, “Computer Vision and Pattern Recognition,” in Proc. of the IEEE Conference on Computer Vision and Pattern recognition, 1997, pp. 193-199.
[9] I. B. Ozer, W. Wolf, “Human detection in compressed domain,” in Proc. of IEEE International Conference on Image Processing, 2001, pp. 274-277, vol. 3.
[10] C. Stauffer, W.E.L. Grimson, “Learning patterns of activity using real-time tracking,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 22, pp.747-757, 2000.
[11] B. Liu, H. Zhou, “Using object classification to improve urban traffic monitoring system,” in Proc. of IEEE International Conference on Neural Networks and Signal Processing, 2003, pp. 1155 - 1159. vol. 2.
[12] I. Haritaoglu, D. Harwood and L. S. Davis, “W4: Real-Time Surveillance of People and Their Activities”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 809-830, 2000.
[13] F. Bobick, J.W. Davis ,” The recognition of human movement using temporal templates” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 23, no. 3, pp. 257-267, 2001.
[14] C. R. Wren, A. Azarbayejani, T. Darrell, and A. Pentland ,”Pfinder: Real-time tracking of the human body,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 19, no 7, pp. 780-785, 1997.
[15] S. Ju , M. Black and Y. Yaccob, “Cardboard people: a parameterized model of articulated image motion”, in Proc. of the IEEE International Conference on Automatic Face and gesture Recognition, 1996, pp. 38-44.
[16] F. Niu,” View-invariant human activity recognition based on shape and motion features”, in Proc. of the IEEE International Symposium on Multimedia Software Engineering, 2004, pp 546 - 556.
[17] R.T. Collins, A. J. Lipton, T. Kanade, H. Fujiyoshi, D. Duggins, Y. Tsin, D. Tolliver, N. Enomoto, O. Hasegawa, P. Burt and L. Wixson, “A system for surveillance and monitoring,” The Robotics Institute, Carnegie Mellon University, Pittsburgh PA, Tech. Rep. CMU-RI-TR-00-12, 2000.
[18] K. P. Karmann, A. von Brandt, “Moving object recognition using and an adaptive background memory,” in Proc. of Elsevier Science Publishers B.V on Time-Varying Image Processing and Moving Object Recognition, 1990, pp. 289-296.
[19] K. P. Karmann, A. V. Brandt, and R. Gerl, “Moving object segmentation based on adaptive reference images,” in Proc. of Elsevier Science Publishers B.V on Signal Processing V: Theories and Application, 1990, pp. 951-954.
[20] D. Koller, J. Weber, T. Huang, J. Malik, G. Ogasawara, B. Rao, and S. Russell, “Towards robust automatic traffic scene analysis in real-time," in Proc. of the IEEE International Conference on Pattern Recognition, vol. 1,1994, pp. 126-131.
[21] A. Elgammal, R. Duraiswami, D. Harwood and L. Davis, “ Background and Foreground Modeling Using Nonparametric Kernel Density Estimation for Visual Surveillance,” Proceedings of The IEEE, vol. 90, no. 7, pp. 1151- 1163, 2002.
[22] E. Davies, Machine Vision: Theory, Algorithms and Practicalities, Academic Press, 1990
[23] M. Sonka, V. Hlavac and R. Boyle, Image Processing, Analysis, and Machine Vision. Pacific Grove: PWS, 1999.
[24] R. C. Gonzalez and R. E. Woods, Digital Image Processing. New Jersey: Prentice-Hall, 2002.
[25] P.J. Deer, P.W., Eklund and B.D, Norman, “A Mahalanobis Distance Fuzzy Classifier,” in Proc. of the IEEE International Conference on Intelligent Information Systems, 1996, pp. 220-223.
[26] C. E. Erdem, A. M. Tekalp and B. Sankur, “Metrics for performance evaluation of video object segmentation and tracking without ground-truth,” in Proc. of the IEEE International Conference on Image Processing, 2001, pp. 69-72, vol. 2.
[27] I. Haritaoglu, D. Harwood and L.S. Davis;” Ghost: a human body part labeling system using silhouettes,” in Proc. of the IEEE International Conference on Pattern Recognition, vol. 1, 1998, pp. 72-82.
[28] http://ftp.cs.rdg.ac.uk/