本文提出一套單眼視覺伺服系統，包含影像處理與伺服控制兩部份。在影像處理方面，藉由裝設於Pan-Tilt機台上的攝影機所擷取之影像，使用以樣本比對為基礎的辨識法，估測出目標與影像中心的移動量，控制機台追蹤，使目標能維持在影像中心，並適時更新樣本以確保當目標物變化時，仍可有效追蹤。在伺服控制方面，使用卡爾曼濾波器，讓系統估測機台在下一時刻將目標物置於影像中心的旋轉量，控制機台可以預先轉動到定位，讓此視覺伺服系統在追蹤方面更有效率。

A monocular visual servo system for a target with variable shape has been developed in this paper. It consists of two parts: an image-processing unit and a servo control unit. For the image-processing unit, the motion between the target and image center is determined by a template match approach. The image is grabbed by the camera equipped on a Pan-Tilt robot and the robot is controlled to track the target by maintaining the target on the image center. However, the template needs to be updated when the target deforms. For the servo control unit, the movement is estimated by the Kalman filter technique to enhance the tracking performance of the visual servo system.

目錄i
圖索引iv
表索引vii
摘要viii
Abstractix
第一章 緒論1
1.1動機與目的1
1.2文獻回顧3
1.3本文架構5
第二章 影像處理運算6
2.1灰階6
2.2差分濾波器7
2.3影像旋轉11
第三章 影像匹配概論15
3.1樣本匹配15
3.2核密度估計19
3.3 Bhattacharyya Coefficient21
第四章 目標追蹤24
4.1方向碼直方圖25
4.2密度函數比對30
4.3整合特徵比對與更新樣本34
4.4區域搜尋38
第五章 攝影機座標與機台控制42
5.1攝影機模型42
5.2 PAN-TILT機台45
5.3卡爾曼濾波器48
5.4應用卡爾曼濾波器於PAN-TILT機台54
第六章 實驗與分析56
6.1軟硬體設備概述56
6.2實驗與驗證58
6.2.1平移運動58
6.2.2平移與旋轉運動59
6.2.3遮蔽63
6.2.4變形65
6.2.5模糊68
6.2.6卡爾曼濾波器於機台控制69
第七章 結論與未來展望76
參考文獻78
附錄82

[1]W. T. Park, and J. Hill, “Real time control of a robot with a mobile camera,” Proc. 9th ISIR, Washington, D.C., p 233-246, 1979.
[2]K. Mitani, and H. Saji, “Robust template matching by using variable size block division,” Proceedings of SPIE - The International Society for Optical Engineering, v 6051, Optomechatronic Machine Vision , p 60510X, 2005.
[3]Y. Li, J. Liu, J. Tian, and H. Xu,”A fast rotated template matching based on point feature,” Proceedings of SPIE - The International Society for Optical Engineering, v 6043 II, MIPPR 2005: SAR and Multispectral Image Processing, p 60431P, 2005.
[4]I. Sobel, “Neighborhood coding of binary images for fast contour following and general array binary processing,” Computer Graphics and Image Processing, v 8, p 127-135, 1978.
[5]J. Canny, “A computational approach to edge detection,” IEEE Trans. on Pattern Analysis and Machine Intelligence, v 8, n 6, p 679-698, 1986.
[6]F. Ullah, S. Kaneko, and S. Igarashi, “Orientation code matching for robust object search,” IEICE Transactions on Information and Systems, v E84-D, n 8, Software Agents and Its Applications/Image Recognition and Understanding, p 999-1006, August 2001.
[7]F. Ullah, S. Kaneko, and S. Igarashi, “Using orientation codes for rotation-invariant template,” Pattern Recognition, v 37, n 2, p 201-209, February 2004.
[8]T. Urano, S. Kaneko, and T. Tanaka, “Fast and robust rotation-invariant search by using orientation code difference histogram,” Proceedings of SPIE - The International Society for Optical Engineering, v 6051, Optomechatronic Machine Vision, p 605101, 2005.
[9]D. Miyamotol, S. Nara, S. Takahashi, and S. Kaneko, “Tracking control system based on orientation code matching,” Proceedings of SPIE - The International Society for Optical Engineering, v 6052, Optomechatronic Systems Control, p 60520D, 2005.
[10]Z. Yin, and R. Collons, “On-the-fly object modeling while tracking,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07, p 4270262, 2007.
[11]C. Harris and M. Stephens, “A Combined Corner and Edge Detector,” Proceedings of the Fourth Alvey Vision Conference, p 147-151, 1988.
[12]D. Kragic, and H. I. Christensen, “Cue Integration for Visual Servoing,” IEEE Transactions on Robotics and Automation, v 17, n 1, p 18-27, February 2001.
[13]R. Li, B. Zeng, and M. Liou, “A new three-step search algorithm for block motion estimation,” IEEE Trans. CASVT, v 4, n 4, p 438-442, August 1994.
[14]D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, v 2, p 142-149, 2000.
[15]D. S. Park, S. Yoon, and Y. B. Kim, “Robot end-effector recognition using modular neural network for autonomous control,” Proceedings of the International Joint Conference on Neural Networks, v 3, p 2032-2036, 1999.
[16]S. Baek, D. S. Park, J. Cho, and Y. B. Lee, “A robot endeffector tracking system based on feedforward neural network,” Robotics and Autonomous Systems, v 28, n 1, p 43-52, July 1999.
[17]X. Zhang, H. Sun, and Y. Wang, “Integrated intensity, orientation code and spatial information for robust tracking ,” ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications, ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications, p 1842-1846, 2007.
[18]Z. C. Hao, M. Zhu, and W. Liu, “Fast extracting and tracking of target in complex environment,” Jilin Daxue Xuebao (Gongxueban)/Journal of Jilin University (Engineering and Technology Edition), v 36, n 2, p 259-263, March 2006.
[19]M. Rosenblatt, “Remarks on some nonparametric estimates of a density function,” Annals Mathematical Statistics, v 27, n 3, p 832-837, 1956.
[20]E. Parzen, “On estimation of a probability density function and mode,” Annals Mathematical Statistics, v 33, n 3, p 1065-1076, 1962.
[21]A. Bhattacharyya, “On a measure of divergence between two statistical populations defined by probability distributions,” Bull. Calcutta Math. Soc., v 35, p 99-109, 1943.
[22]R. E. Kalman, “A new approach to linear filtering and prediction problems,” Transactions of the ASME, Ser. D., Journal of Basic Engineering, n 82, p 34-45, 1960.
[23]R. P. Denaro and P. V. W. Loomis, “GPS navigation processing and Kalman filtering,” AGARD, n 161, p 11.1-11.9, 1989.
[24]洪敏偉，“水下機械手臂之研究與設計”，國立中山大學機械與機電工程學系研究所碩士論文，民國九十五年六月。