在本論文中，我們利用網格架構的分類技巧達到快速遮蔽物的追蹤。在系統中我們強調遮蔽時的追蹤，透過顏色及空間的資訊，計算遮蔽區域中物體分類的相似度。為了應用於即時追蹤系統，我們採用一個基本的網格架構技巧。將遮蔽區域劃分為等大小的網格，接著針對遮蔽區域裡的每個網格做分類，相較於對每個像素做分類，網格架構的技巧可以節省非常多的運算時間。另外，在效能及準確度的權衡上，追蹤系統可以根據物體的大小，自動選擇最恰當的網格大小，除了節省運算時間，同時也保持相當的準確度。在實驗結果中顯示最大的效能可以節省高達98%的執行時間。同時由實驗數據顯示網格架構分類優於像素級分類，且適用於即時追蹤系統。

A fast object occlusion tracking based on grid scheme classification is presented in this thesis. Occlusion tracking is incorporated into our tracking system, which classifies objects in the occlusion region by measuring similarity using both color information and spatial information. In order to be applicable for real time tracking, a simple grid scheme is proposed. The occlusion region is divided into grids, and we then classify each grid in the occlusion region instead of each pixel so the time complexity can be reduced. In addition, according to object size, the proposed grid scheme automatically selects appropriate grid-size based on a good tradeoff between efficiency and accuracy. Experimental results show that execution time can be saved up to 98%. The grid scheme is much faster than the pixel-level approach, and our tracking system can track objects accurately and efficiently.

[1] Y. Alper, J. Omar, and S. Mubarak, “Object tracking: a survey,” ACM Computing Surveys, vol. 38, 2006.
[2] Q. Wei and D. Schonfeld, “Real-time decentralized articulated motion analysis and object tracking from videos,” IEEE Transactions on Image Processing , vol. 16, pp. 2129-2138, 2007.
[3] Y. Li, J. Yang, R. Wu, and F. A. G. F. Gong, “Efficient object tracking based on local invariant features,” International Symposium on Communications and Information Technologies, pp. 697-700, 2006.
[4] P. Kumar, S. Ranganath, K. Sengupta, and H. A. W. H. Weimin, “Cooperative multitarget tracking with efficient split and merge handling,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 16, pp. 1477-1490, 2006.
[5] J. Hao, S. Fels, and J. J. Little, “A linear programming approach for multiple object tracking,” IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, Minnesota, pp. 1-8, 18-23 June, 2007.
[6] P. Jiyan and H. Bo, “Robust occlusion handling in object tracking,” IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, Minnesota, pp. 1-8, 18-23 June, 2007.
[7] A. D. Jepson, D. J. Fleet, and T. F. El-Maraghi, “Robust online appearance models for visual tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, pp. 1296-1311, 2003.
[8] W. Shu-Peng and J. Hong-Bing, “A new appearance model based on object sub-region for tracking,” International Conference on Wavelet Analysis and Pattern Recognition, Beijing, China, pp. 929-932, 2-4 Nov., 2007.
[9] Z. Wen and Z. Cai, “A robust object tracking approach using mean shift,” Third International Conference on Natural Computation, pp. 170-174, 2007.
[10] Y. Shi, J. Konrad, and W. C. Karl, “Multiple motion and occlusion segmentation with a multiphase level set method,” Visual Communications and Image Processing, pp. 189-197, 2004.
[11] A. Bartesaghi and G. Sapiro, “Tracking of moving objects under severe and total occlusions,” IEEE International Conference on Image Processing, Minnesota Univ., USA, pp. I-301-4, 11-14 Sept., 2005.
[12] J. D. Jackson, A. J. Yezzi, and S. Soatto, “Tracking deformable moving objects under severe occlusions,” IEEE Conference on Decision and Control, Bahamas, vol. 3, pp. 2990-2995, Dec. 14-17, 2004.
[13] 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, pp. 1151-1163, 2002.
[14] H. Wang and D. Suter, “A consensus-based method for tracking: modelling background scenario and foreground appearance,” Pattern Recognition, vol. 40, pp. 1091-1105, 2007.
[15] R. Rosales and S. Sclaroff, “3D trajectory recovery for tracking multiple objects and trajectory guided recognition of actions,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Colorado, USA, vol. 2, p. 117-123, 23-25 June, 1999.
[16] S. K. Weng, C. M. Kuo, and S. K. Tu, “Video object tracking using adaptive Kalman filter,” Journal of Visual Communication and Image Representation, vol. 17, pp. 1190-1208, 2006.
[17] S. J. McKenna, S. Jabri, Z. Duric, A. Rosenfeld, and H. Wechsler, “Tracking groups of people,” Computer Vision and Image Understanding, vol. 80, pp. 42-56, 2000.
[18] J. J. Lin, “Efficient image sequence-based point tracking with multiresolution concept and conditional linear filter,” Master Thesis, National Cheng Kung University, Tainan, Taiwan, 2006.
[19] G. Bishop and G. Welch, “An introduction to the Kalman filter,” University of North Carolina at Chapel Hill, Dept. of Computer Science, TR 95-041.