在連續變動的影像平面中，攝影機與環境之相對速度會造成各個影像像素(Pixel)之運動，此種影像像素的運動速度即稱為光流(Optical flow)。以光流為基礎之視覺伺服(Visual servo)方法的主要優點乃在於不需要事先定義目標物特徵，可以適用於較廣泛的定位與追蹤作業。
本文目的為應用視覺伺服技術處理觀看者與環境之3D相對運動問題。將影像當成控制系統的輸入和輸出訊號，以萃取影像間的相對速度資訊。然後使得攝影機可以自動跟隨著運動，保持目標影像不變。

Optical flow, caused by relative motion of the object and the viewer, is the distribution of apparent velocities of brightness pattern in an image. The advantage of the optical-flow-based visual servo method is that feature of the object does not need to be defined or known in advance.
This research plans to build an image servo technique to deal with the problem of 3D relative motion of the viewer and the environment. The images are treated as input and output signals of the control system and are fed back to extract the relative velocity information between contiguous image patterns. Then the video camera will automatically follow the motion to maintain the target image unchanged.

目錄 i
圖索引 iii
表索引 v
摘要 vi
Abstract vii
第一章 緒論 1
1.1 動機與目的 1
1.2 文獻回顧 2
1.3 論文架構 3
第二章 光流系統 5
2.1 光流及影像流的定義 5
2.2 疊代法求解光流 6
2.3 最小平方法求解光流 10
第三章 座標轉換與單位轉換 12
3.1 空間點投影於影像平面之座標轉換　 12
3.2 單位轉換 15
第四章 提升光流精確度方法探討 18
4.1 標準差門檻法 21
4.2 方向特徵門檻法 25
4.3 微分區域擴展法 31
第五章 影像追循實驗 36
5.1 實驗架構 36
5.2 光流適用範圍分析 39
5.3 影像追循實驗 46
第六章 結論與未來展望 57
參考文獻 59
附錄 數位影像單位轉換係數 之求法 62

[1]. Allen, P. K., A. Timcenko, B. Yoshimi and P. Michelman, “Hand-Eye Coordination for Robotic Tracking and Grasping,” in Hashimoto, K. (Ed.), Visual Servoing, World Scientific Publishing Co. Pte. Ltd., pp. 33-69, 1993.
[2]. Cork, P. I., “Visual Control of Robot Manipulators – A Reviews,” in Hashimoto, K. (Ed.), Visual Servoing, World Scientific Publishing Co. Pte. Ltd., pp. 1-31, 1993.
[3]. Feddema, J. T., C. S. G. Lee and P. R. Mitchell, “Feature-based Visual Servoing of Robotic System.” In Hashimoto, K. (Ed.), Visual Servoing, World Scientific Publishing Co. Pte. Ltd., pp. 105-137, 1993.
[4]. Yuh, J. and S. Negahdaripour, “Adaptive Control with Visual Sensing for ROVs in Unstructured Undersea Environments,” International Journal of Robotics and Automation, Vol. 5, No. 1, pp. 32-39, 1990.
[5]. 陳建守，海底無人小艇之光流視覺伺服控制，國立中山大學機械工程研究所碩士論文，中華民國八十五年五月。
[6]. Horn, B. K. P. and B. G. Schunck, “Determining Optical Flow,” Artificial Intelligence, Vol. 17, pp. 185-203, 1981.
[7]. Nagel, H. H., “Displacement Vectors Derived from Second Order Intensity Variations in Image Sequences,” Computer Vision, Graphics and Image Processing, Vol. 21, pp. 85-117, 1983.
[8]. Bartolini, F. and A. Piva, “Enhancement of The Horn and Schunck Optical Flow Algorithm by Means of Median Filters,” Proceedings of IEEE International Conference on Digital Signal Processing, pp. 503-506, 1997.
[9]. Alparone, L., M. Barni, F. Bartolini and R. Caldelli, “Regularization of Optic Flow Estimates by Means of Weighted Vector Median Filtering,” IEEE Transactions on Image Processing, Vol. 8, pp. 1462-1467, 1999.
[10]. Bab-Hadiashar, A. and D. Suter, “Optical Flow Calculation Using Robust Statistics,” Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 988-993,1997.
[11]. Ye, M. and R. M. Haralick, “Optical Flow From A Least-Trimmed Squares Based Adaptive Approach,” Proceedings of IEEE International Conference on Pattern Recognition, Vol. 3, pp. 1052-1055, 2000.
[12]. Lucas, B. and T. Kanade, “An Iterative Image Registration Technique with an Application to Stereo Vision,” Proc. DARPA Image Understanding Workshop, pp. 121-130, 1981.
[13]. Simoncelli, E. P., E. H. Adelson and D.J. Heeger, “ Probability Distributions of Optical Flow,” Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 310-315, 1991.
[14]. 李慧婷，以特徵為基礎之光流計算方法，國立中山大學機械工程研究所碩士論文，中華民國八十八年六月。
[15]. Barni, M., F. Buti, F. Bartolini and V. Cappellini, “A Quasi-Euclidean Norm to Speed Up Vector Median Filter,” IEEE Transactions on Image Processing, Vol. 9, pp. 1704-1709, 2000.
[16]. Sun, S., D. Haynor and Y. Kim, “Motion Estimation Based on Optical Flow with Adaptive Gradients,” Proceedings of IEEE International Conference on Image Processing, Vol. 1, pp. 852-855, 2000.
[17]. Wong, R. Y. and E. L. Hall, “Sequential Hierarchical Scene Matching,” IEEE Transactions on Computers, Vol. 27, No. 4, pp. 359-366, 1978.
[18]. Burt, P. J., C. Yen and X. Xu, “Multi-resolution Flow –through Motion Analysis,” Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 246-252, 1983.
[19]. Lai, S. H. and B. C. Vemuri, “Robust and Efficient Algorithms for Optical Flow Computation,” Proceeding of IEEE International Conference on Computer Vision, pp. 455-460, 1995.
[20]. Wu, Y. T., T. Kanade, J. Cohn and C. C. Li, “Optical Flow Estimation Using Wavelet Motion Model,” Proceedings of IEEE International Conference on Computer Vision, pp. 992-998, 1998.
[21]. Adelson , E.H. and J. R. Bergen, “Spatiotemporal Energy Models for the Perception of Motion,” Journal of the Optical Society of America, A: Optics and Image Science, Vo. 2, pp. 284-299, 1985.
[22]. Heeger, D., “A Model for Extraction of Image Flow,” Proceedings of IEEE Conference on Computer Vision of London, pp. 181-190, 1987.
[23]. Klette, R., “Computer Vision,” Auckland University, New Zealand, http://citr.auckland.ac.nz/teaching/paper/775, pp. 74, 2000.