手語辨識常應用於人機介面中，而移動軌跡是屬其中的一部分，而近年來，有關物體的追蹤研究被廣泛的注意，而多目標追蹤是其中的一個問題。多目標追蹤是利用濾波策略，搭配相關聯的適當物體的資料，用來更新物體追蹤。
三種常見的濾波器分別為卡爾曼濾波器、非察覺型卡爾曼濾波器、粒子濾波器。我們把這三種濾波器應用於追蹤雙手並使用在手發生遮蔽的手語上。卡爾曼濾波器和非察覺型卡爾曼濾波器在手遮蔽時以三種情況的觀測值來分別追蹤手。手發生遮蔽時三種情況為：(a)得到雙手遮蔽時手的中心點，(b)用k均值演算法來得到兩個觀測值，(c)不使用觀測值。
粒子濾波器把手放在一個均勻分布粒子的視窗裡，在追蹤手時讓手在大部分的時間都處在視窗中。粒子的算法分兩種：(a)使用上一態的權重，(b)未使用上一態的權重。最後我們比較三種濾波器在各種不同情況的執行結果並加以討論。

Sign word discrimination has been widely applied in human-computer interface and moving trajectory is part of it. The research for tracking objects has received great attention for the past few years. One of the problems is Multitarget tracking (MTT). MTT uses the filtering scheme which employs associated measurements with proper objects to update object tracking.
The three kinds of the filters are respectively known as the Kalman filter, the unscented Kalman filter, and the particle filter. We apply the three filters to track the two hands and arrange the sign words with the hands occlusion. The Kalman filter and the unscented Kalman Filter use the observations that are classified into three situations to track the hands during the hand occlusion. The three situations are: (a) to obtain the central point when hands happen to occlude; (b) to obtain two observations by k-means algorithm; (c) with no observation.
The Particle filter confines the hand within the window in which particles disperse uniformly. We keep the hand inside the window during the hands tracking for most of the times. The computation of the particle weight is classified into two cases. The two cases are (a) with previous weight; (b) without previous weight. Finally, we compare the results of the three filters in different situations and discuss the results.

1. INTRODUCTION 1
1.1 Motivation 1
1.2 Objective 4
2. METHOD 8
2.1 Kalman Filters 8
2.2 Unscented Kalman Filters 10
2.3 Particle Filters 13
3. EXPERIMENTAL RESULTS 19
3.1 Data sets 19
3.2 Kalman filter and unscented Kalman filter 21
3.3 Particle filter 24
3.4 Comparison 27
4. CONCLUSIONS 32
References 34

[1] P. Kumar, S. Ranganath, K. Sengupta, and H. Weimin,“Cooperative multitarget tracking with efficient split and merge handling,”IEEE Transactions on Circuits and System for Video Technology, vol. 16, no.12, pp. 1477–1490, Dec. 2006.
[2] C. Rasmussen, and G. D. Hager,“Probabilistic data association methods for tracking complex visual objects,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no.6, pp. 560–576, June 2001.
[3] I. J. Cox,“A review of statistical data association techniques for motion correspondence,”International Journal Computer Vision, vol. 10, no.1, pp. 53–66, 1993.
[4] H. T. Nguyen and A. W. M. Smeulders,“Fast occluded object tracking by a robust appearance filter,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, no.8, pp. 1099–1104, 2004.
[5] G. Welch and G. Bishop,“An introduction to the Kalman filter,”July 2006.
[6] D. S. Jang, S. W. Jang, and H. I. Choi,“2D human body tracking with structural Kalman filter,”Pattern Recognition, vol. 35, pp. 2041–2049, 2002.
[7] O. Masoud and N. P. Papanikolopoulos,“A novel method for tracking and counting pedestrians in real-time using a single camera,”IEEE Transactions on Vehicular Technology, vol. 50, no.5, pp. 1267–1278, Sep. 2001.
[8] T. Zhao and R. Nevatia,“Tracking multiple humans in complex situations,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, no. 9, pp. 1208–1221, Sep. 2004.
[9] S. L. Dockstader and A. M. Tekalp,“On the tracking of articulated and occluded video object motion,”Real-Time Imaging, pp. 415–432, Oct. 2001.
[10] S. K. Weng, C. M. Kuo, and S. K. Tu,“Video object tracking using adaptive Kalman filter,”Journal of Visual Communication and Image Representation, pp. 1190–1208, Dec. 2006.
[11] S. J. Julier and J. K. Uhlmann,“A new extension of the Kalman filter to nonlinear systems,”in Proceedings SPIE, vol. 3068, pp. 182–193, 1997.
[12] E. A. Wan and R. Van Der Merwe,“The unscented Kalman filter for nonlinear estimation,”Adaptive Systems for Signal Processing, Communications, and Control Symposium, pp. 153–158, 2000.
[13] N. J. Gordon, D. J. Salmond, and A. F. M. Smith,“Novel approach to nonlinear/non-Gaussian Bayesian state estimation,”in IEE Proceedings-F, vol. 140, pp. 107–113, Apr. 1993.
[14] A. S. Bashi, V. P. Jilkov, X. R. Li, and H. Chen,“Distributed implementations of particle filters,”Information Fusion, Proceedings of the Sixth International Conference of, vol. 2, pp. 1164–1171, 2003.
[15] J. S. Zelek, Topics in Visual Tracking, Particle Filtering, 2007.
[16] B. Zhang, W. Tian, and Z. Jin,“Robust appearance-guided particle filter for object tracking with occlusion analysis,”International Journal of Electronics and Communications, vol. 62, pp. 24–32, 2008.
[17] K. Nummiaro, E. Koller-Meier, and L. V. Gool,“An adaptive color-based particle filter,”Journal of Image and Vision Computing, 2003.
[18] 史文漢、丁立芬，手能生橋，第一冊~第二冊，中華民國聾人協會發行，2004.
[19] A. Almeida, J. Almeida, and R. Araujo,“Real-time tracking of multiple moving objects using particle filters and probabilistic data association,” Journal for Control, Measurement, Electronics, Computing and Communications, vol. 46, Dec. 2005.