臺灣博碩士論文加值系統

現今的追蹤方法主要可以分為兩種：隨機式與決定式。決定式追蹤方法藉由反覆地搜尋，可以在短時間內得到最佳的追蹤結果，但可能會落入區域最大值，而造成追蹤結果錯誤。相對地，隨機式追蹤方法可具有一定的錯誤容忍，但會有較高的複雜度。一般而言，應用於協助教練和足球員分析敵隊進攻之多媒體系統，對於系統複雜度的要求較低。
在本研究中，我們提出一種新的階層式粒子濾波器為基礎之足球運動追蹤系統。本系統之目標在於有效且準確地追蹤足球員。藉由顏色、線條、運動慣性等資訊來追蹤多足球員(移動物體)。同時，也針對遮蔽的情況提供一個有效的處理方法。根據本研究之實驗結果可以發現，特別是在遮蔽的情況發生時，此系統能夠有效追蹤多足球員。

Existing tracking methods can be roughly classified into two categories: stochastic and deterministic. Deterministic methods usually obtain tracking results in a short time by performing searches iteratively. But they may plunge to local maxima, resulting in an error. It is hard to track the target object correctly once an error occurs. Corresponsively, stochastic methods can tolerate some distortions, with higher complexity. Multimedia applications, used to help coaches and players to analyze their competitors’ offensives, have smaller consideration in complexity.
In this study, a new hierarchical particle filter based tracking system for soccer game analysis is proposed. The objective of the proposed system is to track soccer players more efficiently and more precisely. Color, edge, and position information are used to track multiple players (moving objects). Additionally, the proposed system will handle the occlusion problem in an effective manner. Based on the experimental results obtained in this study, the proposed system tracks soccer players more efficiently and correctly, especially when occlusions happen.

TABLE OF CONTENTS

摘 要 i
ABSTRACT ii
ACKNOWLEDGMENTS iii
TABLE OF CONTENTS iv
LIST OF FIGURES vi

CHAPTER 1 INTRODUCTION 1
1.1 Motivation 1
1.2 Survey of Related Researches 3
1.3 Thesis Organization 8

CHAPTER 2 METHODS OF OBJECT TRACKING 9
2.1 Introduction to Tracking Methods 9
2.2 Particle Filter 10
2.3 Mean Shift Algorithm 13
2.4 Cascaded Features 15

CHAPTER 3 PROPOSED TRACKING SYSTEM FOR SOCCER GAME ANALYSIS 19
3.1 System Architecture 19
3.2 Initialization 21
3.3 Hierarchical Particle Filter and Its Features 23
3.4 Occlusion Handling 30

CHAPTER 4 SIMULATION RESULTS 35

CHAPTER 5 DISCUSSIONS AND CONCLUSIONS 58
5.1 Discussions 58
5.2 Conclusions 59

REFERENCES 60

LIST OF FIGURES

Fig. 1.1 Two typical frames derived from broadcast soccer video: (a) close-up, (b) far-view. 3
Fig. 2.1 Illustration of particle filter 11
Fig. 2.2 Representation of general particle filter 13
Fig. 2.3 Presentation of mean shift algorithm [36] 14
Fig. 2.4 Flowchart of mean shift algorithm 15
Fig. 2.5 Flowchart of the cascaded feature structure 16
Fig. 3.1 Flowchart of the proposed system 20
Fig. 3.2 An interested soccer player circled by a rectangle 21
Fig. 3.3 Flowchart of the proposed cascaded feature structure. 23
Fig. 3.4 Representation of object segmentation 24
Fig. 3.5 Representation of deleting four among twenty frames for the position feature... 26
Fig. 3.6 Representation of object motion 28
Fig. 3.7 Representation of eight orientations 28
Fig. 3.8 Presentation of occlusion between teammates and competitors 31
Fig. 3.9 Flowchart of occlusion handling 31
Fig. 3.10 The occlusion representation in four situations 32
Fig. 3.11 The shortest distance between two soccer players 34
Fig. 4.1 The tracking results of the first “Italy versus Ukraine” game sequence using mean shift algorithm 36
Fig. 4.2 The tracking results of the first “Italy versus Ukraine” game sequence using hierarchical particle filter 37
Fig. 4.3 The tracking results of the first “Italy versus Ukraine” game sequence using hybrid system combining mean shift algorithm and particle filter. 39
Fig. 4.4 The tracking results of the first “Italy versus Ukraine” game sequence using our proposed system. 40
Fig. 4.5 Comparison of tracking accuracy of the four methods for first “Italy versus Ukraine” game sequence. 42
Fig. 4.6 The tracking results of “Spain versus France” game sequence using mean shift algorithm 43
Fig. 4.7 The tracking results of “Spain versus France” game sequence using hierarchical particle filter 45
Fig. 4.8 The tracking results of “Spain versus France” game sequence using hybrid system combining mean shift algorithm and particle filter 46
Fig. 4.9 The tracking results of “Spain versus France” game sequence using our proposed system 48
Fig. 4.10 Comparison of tracking accuracy of the four methods for “Spain versus France” game sequence. 50
Fig. 4.11 The tracking results of another “Italy versus Ukraine” game sequence using mean shift algorithm 51
Fig. 4.12 The tracking results of another “Italy versus Ukraine” game sequence using hierarchical particle filter 53
Fig. 4.13 The tracking results of another “Italy versus Ukraine” game sequence using hybrid system combining mean shift algorithm and particle filter. 54
Fig. 4.14 The tracking results of another “Italy versus Ukraine” game sequence using our proposed system. 55
Fig. 4.15 Comparison of tracking accuracy of the four methods for another “Italy versus Ukraine” game sequence 57

[1] I. Bilik and J. Tabrikian, “Target tracking in glint noise environment using nonlinear non-Gaussian Kalman filter,” in Proc. of IEEE Int. Conf. Radar, Apr. 2006, pp. 282-287.
[2] A. Amer, “Voting-based simultaneous tracking of multiple video objects,” IEEE Trans. on Circuits and Schemes for Video Technology, vol. 15, no. 11, pp. 1448-1462, Nov. 2005.
[3] E. Loutas, K. Diamantaras, and I. Pitas, “Occlusion resistant object tracking,” in Proc. of IEEE Int. Conf. on Image Processing, Oct. 2003, pp. 65-68.
[4] J. Li and C. S. Chua, “Transductive inference for color-based particle filter tracking,” in Proc. of IEEE Int. Conf. on Image Processing, Oct. 2003, pp. 14-17.
[5] Z. Q. Wen and Z. X. Cai, “Mean shift algorithm and its application in tracking of objects,” in Int. Conf. on Machine Learning and Cybernetics, Aug. 2006, pp. 4024-4028.
[6] C. Papageorgiou and T. Poggio, “A trainable scheme for object detection,” International Journal of Computer Vision, vol. 38, no. 1, pp. 15-33, 2005.
[7] D. M. Gavrila and V. Philomin, “Real-time object detection for ”smart” vehicles,” in Proc. of IEEE Int. Conf. on Computer Vision, June 1999, pp. 20-25.
[8] E. L. Andrade, J. C. Woods, E. Khan, and M. Ghanbari, “Region-based analysis and retrieval for tracking of semantic objects and provision of augmented information in interactive sport scenes,” IEEE Trans. on Multimedia, vol. 7, no. 6, pp. 1084-1096, Dec. 2005.
[9] P. Rosin, “Thresholding for change detection,” in Proc. of IEEE Int. Conf. on Computer Vision, Jan. 1998, pp. 274-279.
[10] N. Friedman and S. Russell, “Image segmentation in video sequences: a probabilistic approach,” in Proc. of the 13th conf. on Uncertainty in Articial intelligence, Aug. 1997, pp. 175-181.
[11] 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, July 1997.
[12] O. Javed, K. Shafique, and M. Shan, “A hierarchical approach to robust background subtraction using color and gradient information,” in Proc. of IEEE Workshop on Motion and Video Computing, Dec. 2002, pp. 22-27.
[13] M. Cristani, M. Bicego, and V. Murino, “Integrated region- and pixel-based approach to background modeling,” in Proc. of IEEE Workshop on Motion and Video Computing, Dec. 2002, pp. 3-8.
[14] C. Stauffer and E. Grimson, “Adaptive background mixture models for real-time tracking,” in Proc. IEEE Int. Conf. on Computer Vision and Pattern Recognition, 1999, pp. 246-252.
[15] D. S. Lee, J. J. Hull, and B. Erol, “A bayesian framework for Gaussian mixture background modeling,” in Proc. IEEE Int. Conf. on Image Processing, Sept. 2003, pp. 973-976.
[16] A. Elgammal, D. Harwood, and L. Davis, “Non-parametric model for background subtraction,” in Proc. IEEE Int. Conf. on Computer Vision, June 1999, pp. 246-252.
[17] M. Heikkila and M. Pietikainen, “A texture-based method for modeling the background and detecting moving objects,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 28, no. 4, pp. 657-652, Apr. 2006.
[18] Y. Q. Yang, N. Liu, W. Gu, and X. Y. Liang, “An object segmentation approach based on united color models for soccer video,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 28, no. 4, pp. 3960-3963, Aug. 2006.
[19] Q. Wang and J. Liu, “Visual tracking using the kernel based particle filter and color distribution,” in Proc. IEEE Int. Conf. on Neural Networks and Brain, Oct. 2005, pp. 1730-1733.
[20] K. Deguchi, O. Kawanaka, and T. Okatan, “Object tracking by the mean-shift of regional color distribution combined with the particle-filter algorithms,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 3, pp. 506-509, Aug. 2004.
[21] S. McKenna, S. Jabri, Z. Duric, A. Rosenfeld, and H. Wechsler, “Tracking groups of people,” Computer Vision and Image Understanding, vol. 80, pp. 42-56, Oct. 2000.
[22] D. Zhong and S. F. Chang, “An integrated approach for content-based video object segmentation and retrieval,” IEEE Trans. on Circuits and Schemes for Video Technology, vol. 9, no. 8, pp. 1259-1268, Dec. 1999.
[23] N. Paragios and R. Deriche, “Geodesic active contours and level sets for the detection and tracking of moving objects,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 22, no. 3, pp. 266-280, Mar. 2000.
[24] A. Blake and M. Isard, “3D position, altitude and shape input using video tracking of hands and lips,” in Proc. of Computer Graphics, July 1994, pp. 71-78.
[25] D. Terzopoulos and R. Szeliski, “Tracking with Kalman snakes,” in Active Vision, A. Blake and A. Yuille, Eds. Cambridge, MA: MIT Press, 1992, pp. 3-20.
[26] N. Peterfreund, “The velocity snake,” in Proc. of IEEE Workshop on Non-rigid and Articulated Motion, June 1997, pp. 70-79.
[27] N. Peterfreund, “Robust tracking of position and velocity with Kalman snakes,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 21, no. 6, pp. 564-569, June 1999.
[28] K. Hariharakrishnan and D. Schonfeld, “Fast object tracking using adaptive block matching,” IEEE Trans. on Multimedia, vol. 7, no. 5, pp. 853-859, Oct. 2005.
[29] H. Fujiyoshi and A. J. Lipton, ”Real-time human motion analysis by image skeletonization,” in Proc. of IEEE Workshop on Applications of Computer Vision, Oct. 1998, pp.15-21.
[30] R. Polana and R. Nelson, “Low level recognition of human motion,” in Proc. of IEEE Workshop on Motion of Non-Rigid and Articulated Objects, Nov. 1994, pp. 77-82.
[31] C. A. Pau and A. Barber, “Traffic sensor using a color vision method,” in Proc. of SPIE Transportation Sensors and Controls: Collision Avoidance, Traffic Management, and ITS, 1996, vol. 2902, pp. 156-165.
[32] B. Schiele, “Vodel-free tracking of cars and people based on color regions,” in Proc. of IEEE Int. Workshop on Performance Evaluation of Tracking and Surveillance, June 2000, pp. 61-71.
[33] B. Coifman, D. Beymer, P. McLauchlan, and J. Malik, “A real-time computer vision scheme for vehicle tracking and traffic surveillance,” Transportation Researches: Part C, vol. 6, no. 4, pp. 271-288, 1998.
[34] J. Malik and S. Russell, “Traffic surveillance and detection technology development (new traffic sensor technology),” Research Reports, Univ. of California, Berkeley, 1996.
[35] D. S. Jang and H. I. Choi, “Active models for tracking moving objects, ” Pattern Recognition, vol. 33, no. 7, pp. 1135-1146, July 2000.
[36] D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 25, no. 5, pp. 564-575, May 2003.
[37] I. A. Karaulova, P. M. Hall, and A. D. Marshall, “A hierarchical model of dynamics for tracking people with a single video camera,” in Proc. of British Machine Vision Conf., Sept. 2000, pp. 262-352.
[38] Q. Delamarre and O. Faugeras, “3D articulated models and multi-view tracking with physical forces,” Computer Vision and Image Understanding, vol. 81, no. 3, pp. 328-357, Mar. 2001.
[39] S. Ju, M. Black, and Y. Yaccob, “Cardboard people: a parameterized model of articulated image motion,” in Proc. of IEEE Int. Conf. Automatic Face and Gesture Recognition, Oct. 1996, pp. 38-44.
[40] S. A. Niyogi and E. H. Adelson, “Analyzing and recognizing walking figures in XYT,” in Proc. of IEEE Conf. Computer Vision and Pattern Recognition, June 1994, pp. 469-474.
[41] Y. Wu, T. Yu, and G. Hua, “Tracking appearances with occlusions,” in Proc. of IEEE Int. Conf. on Computer Vision and Pattern Recognition, June 2003, vol. 1, pp. I-789-I-795.
[42] W. Hu, T. Tan, L. Wang, and S. Maybank, “A survey on visual surveillance of object motion and behaviors,” IEEE Trans. on Schemes, Man and Cybernetics, Part C: Applications and Reviews, vol. 34, no. 3, pp. 334-352, Aug. 2004.
[43] T. Taki, J. Hasegawa, and T. Fukumura, ”Development of motion analysis scheme for quantitative evaluation of teamwork in soccer games,” in Proc. of IEEE Int. Conf. on Image Processing, Sept. 1996, pp. 815-818.
[44] S. Choi, Y. Seo, H. Kim, and K.S. Hong, “Where are the ball and players? Soccer game analysis with color-based tracking and image mosaick,” in Proc. of Int. Conf. on Image Analysis and Processing, Sept. 1997, pp. 83-91.
[45] S. Iwase and H. Saito, “Tracking soccer players based on homography among multiple views,” in Proc. of SPIE Visual Communications and Image Processing, June 2003, vol. 5150, pp. 283-292.
[46] P. Figueroa, N. Leite, R. M. L. Barros, I. Cohen and G. Medioni, “Tracking soccer players using the graph representation,” in Proc. of Int. Conf. on Pattern Recognition, vol. 4, Aug. 2004, pp. 787-790.
[47] C. Yang, R. Duraiswami, and L. Davis, “Fast multiple object tracking via a hierarchical particle filter,” in Proc. of IEEE Int. Conf. on Computer Vision, vol. 1, Oct. 2005, pp. 212-219.
[48] L. Bretzner, I. Laptev, and T. Lindeberg, “Hand gesture recognition using multi-scale colour features, hierarchical models and particle filtering,” in Proc. of Int. Conf. on Automatic Face and Gesture Recognition, May 2002, pp. 405-410.
[49] L. Fuentes and S. A. Velastin, “Advanced surveillance: from tracking to event detection,” Latin America Transactions, IEEE (Revista IEEE America Latina), vol. 2, no. 3, pp. 1-1, Sept. 2004.
[50] D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 25, no. 5, pp. 564-577 May 2003.
[51] D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proc. of IEEE Int. Conf. on Computer Vision and Pattern Recognition, vol. 2, June 2000, pp. 142-149.
[52] D. Comaniciu and V. Ramesh, “Mean shift and optimal prediction for efficient object tracking,” in Proc. of IEEE Int. Conf. on Image Processing, vol. 3, Sept. 2000, pp. 70-73.
[53] K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image and Vision Computing, vol. 21, pp. 99-110, May 2003.
[54] M. Isard and A. Blake, “Conditional density propagation for visual tracking,” International Journal of Computer Vision, vol. 29, no. 1, pp. 5-28, Aug. 1998.
[55] C. Chang and R. Ansari, “Kernel particle filter for visual tracking,” IEEE Signal Processing Letters, vol. 12, no. 3, pp.242-245, Mar. 2005.
[56] C. Shan, Y. C. Wei, T. N. Tan, and F. Ojardias, “Real time hand tracking by combining particle filtering and mean shift,” in Proc. of IEEE Int. Conf. on Automatic Face and Gesture Recognition, May 2004, pp.669-674.
[57] M. H. Yang, D. Roth, and N. Ahuja, “A snow-based face detector,” in Proc. of Advances in Neural Information Processing Schemes, Dec. 2000, pp. 855-861.
[58] F. Fleuret and D. Geman, “Coarse-to-fine face detection,” International Journal of Computer Vision, vol. 41, no. 1, pp. 85-107, Jan. 2001.
[59] S. Z. Li, Z. Q. Zhang, H. Y. Shum, and H. Zhang, “Floatboost learning for classification,” in Proc. of Neural Information Processing Schemes, Dec. 2002, pp. 993-1000.
[60] P. Viola and M. J. Jones, “Robust real-time face detection,” International Journal of Computer Vision, vol. 52, no. 2, pp.137-154, May 2004.
[61] H. Schneiderman, “Feature-centric evaluation for efficient cascaded object detection,” in Proc. of IEEE Int. Conf. on Computer Vision and Pattern Recognition, vol. 2, July 2004, pp. 29-36.
[62] P. Viola and M. Jones, “Rapid object detection using a boosted cascade of simple features,” in Proc. of IEEE Int. Conf. on Computer Vision and Pattern Recognition, Dec. 2001, pp. 511-518.
[63] B. C. Brandao, J. Wainer, and S. K. Goldenstein, “Subspace hierarchical particle filter,” in Proc. of the Brazilian Symposium of Computer Graphics and Image Processing, Oct. 2006, pp. 194-204.
[64] Y. Kobayashi, et al., “3D head tracking using the particle filter with cascaded classifiers,” in Proc. of the British Machine Vision Conference, Sept. 2006, pp. 37-46.
[65] L. Devroye, L. Gyorfi, and G. Lugosi, “A probabilistic theory of pattern recognition,” Applications of Mathematics, Springer-Verlag Inc., vol. 31, New York, N.Y., 1996, pp. 25-32.
[66] P. Viola and M. Jones, “Robust real-time face detection,” International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154, May 2004.
[67] N. T. Siebel and S. Maybank, “Fusion of multiple tracking algorithms for robust people tracking,” in Proc. of European Conference on Computer Vision, May 2002, pp. 373-387.
[68] C. Hua, H. Wu, Q. Chen, and T. Wada, “A pixel-wise object tracking algorithm with target and background sample,” in Proc. Int. Conf. on Pattern Recognition, vol. 1, 2006, pp. 739-742.
[69] A. G. Perera, et al., “Evaluation of algorithms for tracking multiple objects in video,” in Proc. of IEEE Workshop on Imagery and Pattern Recognition, Oct. 2006, pp. 35-35.
[70] Q. Zhao, J. Kang, H. Tao, and W. Hua, “Part based human tracking in a multiple cue fusion framework,” in Proc. of 2006 Int. Conf. on Pattern Recognition (ICPR), vol. 1, 2006, pp. 450-455.
[71] A. Aristidou, P. Pangalos, and H. Aghvami, “Tracking multiple sports players for mobile display,” in Proc. of 2007 Int. Conf. on Image Processing, Computer Vision, and Pattern Recognition, Nevada, USA, 2007, pp. 53-59.
[72] M. Obata, T. Nishida, H. Miyagawa, and F. Ohkawa, “Target tracking and posture estimation of 3d objects by using particle filter and parametric eigenspace method,” in Proc. of 2007 Int. Conf. on Image Processing, Computer Vision, and Pattern Recognition, Nevada, USA, 2007, pp. 67-72.
[73] S. K. Weng, C. M. Kuo, and S. K. Tu, “Video object tracking using adaptive Kalman filter,” International Journal of Visual Communication and Image Representation, vol. 17, no. 6, pp. 1190-1208, Dec. 2006.