臺灣博碩士論文加值系統

在此論文中，成功地建立一個在開放空間下仍具有高可靠度之即時影像伺服追蹤系統。不僅克服了傳統動態輪廓模型的限制，達成系統可自動偵測並追蹤三度空間中任意形狀物體之目標，還克服了目標物出現在極複雜背景下或被其他物體遮蔽等追蹤問題；而整個系統的執行速度更是少於34豪秒。此追蹤系統包含了運動偵測(motion detection)、以動態輪廓模型為基礎之輪廓截取法(snake-based outline extraction)、混合型追蹤法則(hybrid tracking methodology)以及影像機率資料連結濾波器(visual probabilistic data association filter)。我們的相機是架構在一個雙軸步進馬達平台上，透過影像回饋機制，可以在真實環境中追蹤我們有興趣的目標物。最後，我們做了許多的實驗來驗證此追蹤系統的效能。

This thesis presents a real-time, highly reliable, open field visual tracking system, which can automatically detect an arbitrary-shaped object tin 3-D space and find out its location so that the camera platform can be controlled to keep the target centered in the monitor image. Even if the object goes through highly cluttered environment or is occluded by other objects, the system should not fail to work properly. And the total processing period is less than 34 ms. The overall system consists of a motion detector, a snake-based outline extraction, a hybrid tracking methodology, and a VPDA filter which evolves from Probabilistic Data Association filter (PDA filter). Moreover, the camera is mounted on a two-axis motor platform, allowing alignment of the line of sight with the target of interest. At last, the effective functionality of the visual servo system are confirmed by a series of experiments.

Contents
1. INTRODUCTION…………………………………………………………………1
1-1 MOTIVATION………………………………………………………………..1
1-2 BRIEF LITERATURE REVIEW……………..………………..…………….3
1-3 CONTRIBUTION…………………………………………………………….5
1-4 THESIS ORGANIZATION…………………………………………………..5
2. PRELIMINARY…………………………………………………………………...7
2-1 ACTIVE CONTOUR MODEL………………...…………………………….7
2-2 FUNDAMENTAL IMAGE PROCESSING METHODS………………….....9
2-2-1 Gradient Edge Detection……………………………………………9
2-3-2 Template Matching……………………………………………..11
2-3-3 Gray Scale Image Close Operation…………………………………13
2-3 COORDINATE TRANSFORMATION……………………………………..15
2-4 SYSTEM OVERVIEW……………………………………………………...21
3. TARGET DETECTION AND TRACKING ALGORITHM………………..…24
3-1 ARCHITECTURE OVERVIEW……………………………………………24
3-2 MOTION DETECTION…………………………………………………….25
3-3 SNAKE-BASED OUTLINE EXTRACTION….…………………………...30
3-3-1 Modified Active Contour Method…………….………………..31
3-3-2 Automatic Initialization of Snake……….…………………………36
3-4 OBJECT TRACKING METHODOLOGY……………………………39
3-4-1 Contour Matching………………………………………………..39
3-4-2 Combined with Template Matching……………………………41
3-5 VISUAL PROBABILITY DATA ASSOCIAITON FILTER………………..41
4. TRAJECTORY DESIGN AND VISUAL SERVO……………………………..46
4-1 KALMAN FILTER…………………………………………………………46
4-2 PROBABILISTIC DATA ASSOCIATION FILTER………………………..49
4-2-1 The Validation Gate…………………………………….………49
4-2-2 The Stochastic Model…………………………………………..52
4-3 VISUAL PROBABILISTIC DATA ASSOCIATION FILTER……………...53
4-3-1 Image Feature Similarity Association………………………………54
4-3-2 Persistent Noise Removal………………………………………55
4-4 MOTOR CONTROL………………………………………………………..62
5. EXPERIMENTAL RESULTS………………………………………………...…65
5-1 MODIFIED SNAKE-BASED OUTLINE EXTRACTION………………...65
5-2 COMPARISON WITH DIFFERENT TRACKING ALGORITHM………...70
5-2 APPLY TO TRACK HUMAN HEAD…………………………….. ……….81
6. CONCLUSION AND FUTURE WORK………………………………………..89
REFERENCE……………………………………………………………………….91

References
[1] M. Kass, A. Witkin, and D. Terzopoulos, “Snake: Active Contour Models,” Int. J. Comput. Vis., Vol. 1, pp. 321-331,1987
[2] D.J. Williams and M. Shah. “A Fast Algorithm for Active Contours and Curvature Estimation,” CVGIP: Image Understanding, Vol. 55, No. 1, pp. 14-26, Jan. 1992.
[3] C. Xu and J.L.Prince, "Snakes, Shapes, and Gradient Vector Flow", IEEE Trans. Image Processing, Vol. 7, No. 3, pp. 359369, 1998.
[4] J. Denzeler, and H. Niemann, “A Two Stage Real-Time Object Tracking System,” In Pavesi'c et al.
[5] Isard M. and Blake A., “CONDENSATION─ Conditional Density Propagation for Visual Tracking,” Int. J. Computer Vision, pp. 1-36, 1998.
[6] Isard M. and Blake A., “A Mixed-State CONDENSATION Tracker With Automatic Model-Switching,” Proc. 6th Int. Conf.. Computer Vision, pp. 107-112, 1998.
[7] Isard M. and Blake A., Active Contours: The Application of Techniques From Graphics, Vision, Control Theory and Statistics to Visual Tracking of Shapes in Motion, Spring Inc., 1998
[8] Arsenio A. and Santos-Victor J., “Robust Visual Tracking by an Active Observer,” Int. Conf. on Intelligent Robots and Systems, Vol. 3, pp. 1342-1347, 1997
[9] J. Denzeler and H. Niemann, “ A New Energy Term Combining Kalman Filter and Active Contour Models for Object Tracking,” Machine Graphics Vision, Vol. 5(1/2), pp. 157-165, 1996
[10] Natan Peterfreund, “Robust tracking of Position and Velocity with Kalman Snakes,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 21, No. 6, pp. 564-569, June 1999.
[11] C. Xu and J. L. Prince, “Generalized Gradient Vector Flow External Forces for Active Contours,” Int. J. on Signal Processing, Vol. 71(2), pp. 131-139, Dec. 1998.
[12] C. Xu and J.L. Prince, “Gradient Vector Flow: A New External Force for Snake,” IEEE. Proc. Conf. on Computer Vision and Pattern Recognition, pp. 66-71, 1997
[13] S.W. Zucker, C. David, A. Dobbins, and L. Iverson, “The organization of curve detection: Coarse tangent fields and fine spline coverings,” In Second Int. Conf. on Computer Vision, Tampa, Florida, p.568-577, 1988
[14] Yalamanchili, S., W. N. Martin, and J. K. Aggarwal, “Extraction of moving object description via differencing,” CGIP, Vol. 18, pp. 188-201, 1982.
[15] Murray, D. and A. Basu, “Motion Tracking with an Active Camera,” IEEE Trans. Pattern Analysis and Intelligence, vol. 16, no.5, pp. 449-459, May 1994.
[16] Lee, K. W., S. W. Ryu, S. J. J. Lee, and K. T. Park, “Motion based object tracking with mobile camera,” Electronics letters 5th, vol. 34, no.3, Feb. 1998.
[17] Peacock, A. M., S. Matsunaga, D. Renshaw, J. Hannah, and A. Murray, “Reference Block Updating When Tracking with Block Matching Algorithm,” Electronics letters 17th, Vol. 36, no.34, pp. 309-310, Feb. 2000.
[18] Haworth, C., A. M. Peacock, and D. Renshaw, “Performance of Reference Block Updating Techniques when Tracking with the Block Matching Algorithm,” Proc. Int. Conf. Image Processing, Vol. 1, pp.365-368, 2001.
[19] Birchfield, Stan, “An Elliptical Head Tracker,” 31st Asilomar Conference on Signals, Systems, and Computers, November 1997.
[20] Birchfield, Stan, “Elliptical Head Tracking Using Intensity Gradiants and Color
Histograms,” IEEE Conf. on Computer Vision and Pattern Recognition, Santa Barbara, California, July 1998.
[21] Lipton, A., H. Fujiyoshi, and R. Patil, “ Moving Target Classification and Tracking from Real-Time Video,” Proc. Of the DARPA Image Understanding Workshop, 1998
[22] Brunelli, R. and T. Poggio, “Template Matching: Matched Spatial Filters And Beyond,” MIT AI Memo 1549, July 1995.
[23] Lew, M.S., N. Sube, T.S. Huang, “Improving visual matching,” IEEE Conference on Computer Vision and Pattern Recognition, Vol. 2, pp. 58 —65, 2000.
[24] Lewis, J. P. “Fast Normalized Cross-Correlation,” In Vision Interface, 1995.
[25] Chen, Y. S., Y. P. Hung, and C. S. Fuh, “A Fast Block Matching Algorithm Based on the Winner-Update Strategy,” Proc. of the 4th Asian Conf. on Computer Vision, vol. 2, pp. 977-982, Jan. 2000.
[26] J. W. Lee, M. S. Kim, and I. S. Kweon, “A Kalman Filter Based Visual Tracking Algorithm for an Object Moving in 3D,” IEEE Int. Conf. Intelligent Robots and Systems 95., vol. 1, pp. 342-347, 1995.
[27] Arthur Gelb, Applied Optimal Estimation, the M.I.T. Press, 1974
[28] Bar-Shalom, Y., E. Tse, “Tracking in a Cluttered Environment with Probabilistic Data Association,” Automatica, Vol. 11, pp. 451-460, 1975.
[29] Bar-Shalom, Y., T. E. Fortmann, Tracking and Data Association, Academic Press, 1988.
[30] Colegrove, S. B., and S. J. Davey, “The Probabilistic Data Association Filter with Multiple Nonuniform Clutter Regions,” IEEE Int. Radar Conf., pp. 65-70, 2000.
[31] Sarachik, K. B., “The Effect of Gaussian Error in Object Recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 19, No. 4, pp. 289-301, 1997.
[32] Boshra, M., and B. Bhanu, “Predicting Performance of Object Recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, No. 9, pp. 956-969, 2000.
[33] Liu, D. and L. C. Fu, “Target Tracking in an Environment of Nearly Stationary and Biased Clutter,” IEEE Int. Conf. on Intelligent Robots and Systems, Vol. 3, pp. 1358-1363, 2001.
[34] Chen, J., H. Leung, T. Lo, J. Litva, M. Blanchette, “A Modified Probabilistic Data Association in a Real Clutter Environment,” IEEE Trans. Aerospace and Electronic Systems, Vol. 32, No. 1, pp. 300-313, 1996.
[35] Ridler, T. W., S. Calvard, “Picture Thresholding using an Iterative Selection Method,” IEEE Trans. Systems, Man and Cybernetics, Vol. 8, No. 8, pp. 630-632, 1978.
[36] Trussell, H. J., “Comments on Picture Thresholding using an Iterative Selection Method,” IEEE Trans. Systems, Man and Cybernetics, Vol. smc-9, No. 5, pp. 311, 1979.
[37] Magid, A., S. R. Rotman, and A. M. Weiss, “Comment on Picture Thresholding using an Iterative Selection Method,” IEEE Trans. Systems, Man and Cybernetics, Vol. 20, no. 5, Sep/Oct 1990.
[38] Liu, D., “Real-Time Visual Tracking in Cluttered Environment with a Pan-Tilt Camera,” Master Thesis, Dept. of Electronical Eng., National Taiwan University, 2001.
[39] Kau, T. K., “A Robust Visaul Servo Based Headertracker with Auto-zooming in Cluttered Environment,” Master Thesis, Dept. of Electronical Eng., National Taiwan University, 2002.
[40] Richard M. M., Li Z. and Sastry S. S., A Mathematical Introduction to Robotic Manipulation, CRC Press Inc, 1994
[41] Haralick, R. M. and L. G. Shapiro, Computer and Robot Vision, Vol.1, Addison-Wesley Inc., 1992.
[42] William K. Pratt, Digital Image Processing, 3rd Edition, Wiley Inc., 2001