在本論文中，為了機器的視覺追蹤而建立了一個結合視覺部分和控制部分的影像追蹤系統。我們提出一個影像處理的方法來建立視覺部分，在這影像處理的方法中，是利用兩張由攝影機拍攝的影像來進行影像減算，以便找出影像中有物體移動的區域，將此移動區域對應到第二次拍攝的影像上，再對此區域進行灰階色調與邊界強度的兩種分割。於分割完之後，以兩種分割為考量進行形態的濾除，而概略地找出所要追蹤的目標物。
　　在獲得目標物的位置之後，我們可以計算出位置與中心點的偏差，而將此偏差用來當作控制部分的輸入命令。因此，我們發展自調的模糊控制系統用以傳動攝影機，自調的部分係以系統識別的架構來建立並且以最深梯度法來訓練歸屬函數中的參數值，目的是為了讓模糊控制系統能在不同的情況下都能更強健。此外，視覺部分和控制部分的軟體兩者都是建立在個人電腦的環境下，並且硬體彼此間都是依靠介面卡來溝通。
　　最後，在不同的實例上進行實驗，證明了整個影像追蹤系統能夠在不同的狀況下運作良好，即使是受到一般的干擾。

　　In this thesis, an image tracking system combining visual part and control part is established for visual tracking of robot. We propose a method of image processing to build the visual part. In the method of image processing, two images captured by camera will process image subtraction to find the region that objects are in motion. The motion region corresponding onto the second captured image is processed for the segmentations of gray hues and edge intensity. After segmentations, we implement morphological filter on both segmentations to find the tracked object sketchily.
　　After obtaining the position of the tracked object, we can obtain the information of position deviation and use it as the input command for control part. Therefore, the self-tuning Fuzzy control system is utilized as the motion control part for driving camera. Self-tuning part is built in system identification structure and adopts the steep descent algorithm for training system parameters in the membership functions. The purpose is to make Fuzzy control system be more robust under different conditions. Besides, the software of visual part and motion control part are both built in the environment of PC and mutual communications of the hardware depend on interface cards.
　　Finally, experimenting on different cases proves that the whole image tracking system is able to work well under different conditions, even with common noises.

[1] Y. Wu and T.S. Huang, “Nonstationary Color Tracking for Vision-Based Human-Computer Interaction,” IEEE Transactions on Neural Networks, Vol. 13, No. 4, July 2002, pp. 948-960.
[2] Y. Ricquebourg and P. Bouthemy, “Real-Time Tracking of Moving Persons by Exploiting Spatio-Temporal Image Slices,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 8, August 2000, pp. 797-808.
[3] N. Paragios and R. Deriche, “Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 3, March 2000, pp. 266-280.
[4] D. Murray and A. Basu, “Motion Tracking with An Active Camera,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 16, No. 5, May 1994, pp. 449-459.
[5] Z. Su and K. Khorasani, “A Neural-Network-Based Controller for a Single-Link Flexible Manipulator Using the Inverse Dynamics Approach,” IEEE Transactions on Industrial Electronics, Vol. 48, No. 6, December 2001, pp. 1074-1086.
[6] M. Kinouchi, I. Hayashi, N. Iwatsuki, K. Morikawa, J. Shibata, and K. Suzuki, “Application of Fuzzy PI Control to Improve the Positioning Accuracy of A Rotary-Linear Motor Driven by Two-Dimensional Ultrasonic Actuators,” ELSEVIER Microprocessors and Microsystems, Vol. 24, 2000, pp. 105-112.
[7] E.-W. Huang and W.-G. Yau, “An Edge Based Visual Tracking for Target within Complex Environment,” Proceedings of the American Control Conference, Vol. 3, 28-30 June 2000, pp. 1993-1997.
[8] R. Canals, A. Roussel, J.-L. Famechon, and S. Treuillet, “A Biprocessor-Oriented Vision-Based Target Tracking System,” IEEE Transactions on Industrial Electronics, Vol. 49, No. 2, April 2002, pp. 500-506.
[9] M. Schwarzinger, T. Zielke, D. Noll, M. Brauckmann, and W. von Seelen, “Vision-Based Car-Following: Detection, Tracking, and Identification,” Proceedings of the Intelligent Vehicles '92 Symposium., Vol. , No. , 29 June-1 July 1992, pp. 24-29.
[10] M. Sonka, V. Hlavac and R. Boyle, Image Processing, Analysis, and Machine Vision Second Edition, PWM Publishing, 1999.
[11] R. Jain, R. Kasturi and B. G. Schunck, Machine Vision, First Edition, McGraw-Hill, 1995.
[12] Y. Zhong, A. K. Jain and M. P. Dubuisson-Jolly, “Object Tracking Using Deformable Templates,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 5, May 2000, pp. 544-549.
[13] G.-J. Wang, C.-T. Fong and K.-J. Chang, “Neural-Network-Based Self-Tuning PI Controllers for Precise Motion Control of PMAC Motors,” IEEE Transactions. on Industrial Electronics, Vol. 48, No. 2, April 2001, pp. 408-415.