在視頻監控中異常運動物體偵測是一個重大議題，透過電腦視覺分析視頻中的行人與車輛，進而判斷是否異常。傳統的異常偵測方法是透過事先定義異常事件規則來判斷，但真實世界的情況較為複雜，無法每一種都事先定義。有鑑於以上方法的缺點，我們設計一套學習模式，不需要事先定義異常規則，自動檢測異常事件與物體。
本論文提出一個可應用於在各種環境中,即時自動偵測出異常運動物體的方法。首先偵測場景上的運動物體，我們利用高斯混合模型偵測前景，並透過陰影濾除消除前景陰影，再利用Blobs檢測出運動物體。我們提出一個結合卡爾曼濾波器的改良式Mean Shift演算法對這些運動物體進行追蹤，最後使用卡爾曼濾波器平滑軌跡資料。
在異常運動物體偵測的程序中，將收集到的軌跡資訊，利用自我組織增量神經網路將這些軌跡資訊進行學習，學習出一個正常軌跡模組，學習時間約7到55秒。之後它作為判斷運動物體是否正常的依據，反之即異常。自我組織增量神經網路具有抗雜訊等優點。
實驗的部份我們針對不同環境進行分析，如學校廣場、馬路、單行道。我們提出的方法可以正確偵測到異常物體，在學校廣場的準確性是100％，在馬路是98.3％，在單行道是98.8％，且整體執行時間很短，約0.033到0.067秒，達到即時偵測。

Abnormal moving objects detection is an essential issue for video surveillance. In order to judge whether the behavior of objects is abnormal, such as pedestrians walk back and forth, walk across the street, or scooters drive the wrong way, the main method is through computer vision technique to analyze objects as pedestrians, cars, and so on in video. Traditional abnormal moving objects detection aims at particular circumstances or requirement to predefine particular detection rules which the application of abnormal moving objects detection is restricted. Besides, if numerous abnormal moving objects are detected at the same time, surveillance system is overloaded with operation. Owing to this reason, in this paper, we expect to design a set of learning model which does not predefine abnormal rules and can detect a variety of abnormal moving objects automatically in different environments.
To achieve the above goal, the first thing is to detect the moving objects in video. The proposed method in this paper utilizes Gaussian Mixture Model (GMM) to detect foreground objects and remove shadows of objects by shadow removal. Then, adoptive mean shift algorithm with Kalman filter is proposed to track these moving objects. Finally, Kalman filter is used to smooth trajectory.
After collecting the trajectories of moving objects, abnormal moving object detection process proceeds. At first, for this trajectory information, take advantage of Self-Organizing Incremental Neural Network (SOINN) to learn and build a normal trajectory model which is a foundation to determine whether follow-up moving objects are abnormal. The average learning time is 7 to 55 seconds.
The experiment monitors and analyzes different circumstances, such as School campus, roads, and one-way street. The system based on the proposed method can detect abnormal moving objects with the accuracy 100% in school campus, 98.3% in roads, and 98.8% in one-way street. The overall execution time is short and about 0.033 to 0.067 seconds, and it can be executed in real-time.

中文摘要 i
Abstract ii
致謝 iv
Table of Contents v
List of Figures vi
List of Tables viii
Chapter 1 Introduction 1
1.1 Overview 1
1.2 Motivation 1
1.3 System Description 2
1.4 Thesis organization 3
Chapter 2 Related Works 4
Chapter 3 Moving Object Detection and Tracking 10
3.1 Foreground Detection 10
3.1.1 Background model 10
3.1.2 Shadow Removal 13
3.2 Moving Object Tracking 15
3.2.1 Moving Object Detection and Blobs Tracking 15
3.3 Multiple Objects Tracking with Occlusion Handling 19
3.3.1 Kalman Filter 20
3.3.2 Mean Shift Algorithm 21
3.3.3 Proposed Modified Mean Shift Method 23
3.4 Handling of a Missed Tracking Object 26
3.5 Trajectory Post Processing 28
Chapter 4 Abnormal moving object detection 30
4.1 Trajectory Feature Extraction 30
4.2 Self-Learning Using SOINN 32
4.3 Abnormal moving object detection 39
Chapter 5 Experimental Results and Discussions 42
5.1 Experimental Setup 43
5.2 Results of Moving Object Tracking 44
5.3 The Results of SOINN Learning Trajectory 51
5.4 The Results of Abnormal Moving Object detection 53
Chapter 6 Conclusions and Future Works 61
6.1 Conclusions 61
6.2 Future Works 62
References 63

[1]S. Furao, and O. Hasegawa, “An Incremental Network for Online Unsupervised Classification and Topology Learning,” Neural Networks, vol. 19, no. 1, pp. 90-106, 2006.
[2]F. Bremond, E. Corvee, J. L. P. Vilchis, and M. Thonnat, “CARETAKER Project.” Online] Available: http://www-sop.inria.fr/members/Francois.Bremond/topicsText/caretakerProject.html (accessed on May 25, 2011).
[3]E. Ricci, and G. Zen, “Earth Mover's Prototypes: A Convex Learning Approach for Discovering Activity Patterns in Dynamic Scenes,” in Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI, USA, pp. 3225 - 3232, Jun. 2011.
[4]Y. Ivanov, and A. Bobick, “Recognition of Visual Activities and Interactions by Stochastic Parsing,” IEEE Transactions on Pattern Analysis Machine Intelligence, vol. 22, no. 8, pp. 852-872, 2000.
[5]D. Minnen, I. Essa, and T. Starner, “Expectation Grammars: Leveraging High-Level Expectations for Activity Recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Madison, WI, USA, vol. 2, pp. 626-632, Jun. 2003.
[6]“UCSD anomaly detection dataset.” Online] Available: http://www.svcl.ucsd.edu/projects/anomaly/dataset.html (accessed on April 4, 2011).


[7]A. Galata, N. Johnson, and D. Hogg, “Learning Variable Length Markov Models of Behavior,” Computer Vision and Image Understanding, vol. 81, no. 3, pp. 398-413, 2001.
[8]N. Johnson, and D. Hogg, “Representation and Synthesis of Behavior Using Gaussian Mixtures,” Image and Vision Computing, vol. 20, pp. 889-894, 2002.
[9]C. Stauffer, and W. E. L. Grimson, “Learning Patterns of Activity Using Real-Time Tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 747-757, 2000.
[10]D. Makris, and T. Ellis, “Path Detection in Video Surveillance,” Image Vision and Computation, vol. 20, no. 12, pp. 895-903, 2002.
11]D. Makris, and T. Ellis, “Learning Semantic Scene Models from Observing Activity in Visual Surveillance,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 35, no. 3, pp. 397-408, 2005.
[12]C. Piciarelli, and G. L. Foresti, “On-Line Trajectory Clustering for Anomalous Events Detection,” Pattern Recognition Letters, vol. 27, no. 15, pp. 1835-1842, 2006.
[13]F. Jiang, Y. Wu, and A. K. Katsaggelos, “Abnormal Event Detection Based on Trajectory Clustering by 2-Depth Greedy Search,” in Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, Las Vegas, NV, USA, pp. 2129-2132, Mar. 2008.
[14]F. Jiang, Y. Wu, and A. K. Katsaggelos, “Abnormal Event Detection from Surveillance Video by Dynamic Hierarchical Clustering,” in Proceedings of 2007 IEEE International Conference on Image Processing, San Antonio, TX, USA, vol. 5, pp. 145-148, Sep. 2007.
[15]C. T. Hsieh, S. B. Hsu, C. C. Han, and K. C. Fan, “Abnormal Event Detection Using Trajectory Features,” Journal of Information Technology and Applications, vol. 5, no. 1, pp. 21-27, 2011.
[16]C. Piciarelli, and C. Micheloni, “Trajectory-Based Anomalous Event Detection,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 18, no. 11, pp. 1544-1554, 2008.
[17]P. KaewTraKulPong, and R. Bowden, “An Improved Adaptive Background Mixture Model for Real-Time Tracking with Shadow Detection,” in Proceedings of 2nd European Workshop on Advanced Video Based Surveillance Systems (AVBS), London, UK, pp. 149-158, September 2001.
[18]R. Cucchiara, C. Grana, M. Piccardi, and A. Prati, “Detecting Moving Objects, Ghosts, and Shadows in Video Streams,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 10, pp. 1337-1342, 2003.
[19]“Open Source Computer Vision Library.” Online] Available: http://www.intel.com/technology/computing/opencv/ (accessed on August 14, 2011).
[20]F. Fazal, E. Guraya, PY Bayle, and F. A. Cheikh, “People Tracking via a Modified CAMSHIFT Algorithm,” in Proceedings of DCABES 2009, Wuhan, China, Oct. 2009.
[21]F. Shimin, G. Qing, X. Sheng, and T. Fang, "Human Tracking Based on Mean Shift and Kalman Filter," in Proceedings of the 2009 International Conference on Artificial Intelligence and Computational Intelligence (AICI'09), Chengdu, China, vol. 3, pp. 518-522, Nov. 2009.
[22]G. Bishop and G. Welch, “An Introduction to the Kalman Filter,” in Universityy of Norh Carolina SIGGRAPH 2001 course notes, North Carolina, Course 8, 2001.
[23]G. R. Bradski, “Computer Vision Face Tracking for Use in a Perceptual User Interface,” Intel Technology Journal, vol. 2, no.2, pp. 12-21, 1998.
[24]V. Papadourakis, and A. Argyros, “Multiple Objects Tracking in the Presence of Long-Term Occlusions,” Computer Vision and Image Understanding, vol. 114, no. 7, pp. 835-846, 2010.
[25] S. Furao, T. Ogura, and O. Hasegawa, “An Enhanced Self-Organizing Incremental Neural Network for Online Unsupervised Learning,” Neural Networks, vol. 20 no. 8, pp. 893-903, 2007.