臺灣博碩士論文加值系統

仿生型自主式水下載具之視覺運動估測系統必須能提供外界物體運動估測功能，以作為選擇行為模式之依據。針對此需求，本文提出一個視覺運動估測系統，採用現行光流法及立體成像法，配合自行推導之運動量計算演算法及物體測距。
擷取序列影像後先進行像質改善、分割，接下來視訊訊號處理包含光流法、運動量計算、攝影機校正及立體成像法四部分。像質改善部分採高斯低通濾波；分割部分則考量動態影像特性採序列影像相減；在光流計算方面，採用Horn-Schunck法及SOAD兩種方法；運動量計算部分，將總光流場分解為物體平移、旋轉及距離變化相關三種成分，建立運動參數求取演算法，並使用迴圈疊代，以獲得物體運動量最佳估算值。攝影機校正採用不考慮攝影機物理模型之校正法，結合立體成像法求取空間中物體位置及距離。
視覺程式之驗證採模擬序列影像，針對攝影機及背景靜止而僅物體運動之情形，考慮不同運動狀態、型式及大小，利用不同光流法進行運動估測，並進行攝影機校正及距離量測之驗證。
驗證結果顯示，光流法中以SOAD較佳，物體運動量參數之誤差最大為6%，使用三次攝影機校正可使校正準確度高達95%以上，且在有效範圍內，物體距離量測之誤差也在3%以下。而應用實例中，針對物體連續運動的序列影像實施運動估測，結果顯示本文所建構視覺處理流程確具可行性。
關鍵字：自主式水下載具、仿生學、水下機器人視覺、光流、攝影機校正、立體成像法、運動估測。

This paper presents a visual motion estimation system for biomimetic AUVs. To provide the function of object motion estimation through a machine vision system, after grabbed the sequences of video images, preceded processing is image improvement and segmentation. For image improvement, the Gaussian filtering is applied. Segmentation is when performed by using subtraction operations.
After the preceded processing, the scheme is composed of the optical flow analysis, movement calculation of moving objects camera calibration and stereo imaging. We choose two optical flow algorithms (Horn-Schunck method and SOAD) to calculate optical flow. For optimal movement calculation, the motion parameter algorithm is established, the optical flow is divided into object translation, rotation and distance change. Using a mathematical model for camera calibration, and combine with stereo imaging to obtaining the position and distance of the object in space.
To put programs to the proof, we make up the synthetic image pair. Use two optical flow methods to calculate the motion estimated. We considered different type and degree of movement on condition that the camera and background are still and only one object moving. To grabbing difference distance calibration image to calculate the calibration coefficients, and calculate the degree of accuracy of the coefficients. Last of all, calculate motion estimated for the real sequence images.
Results suggest that: (1) SOAD is batter than Horn-Schunck method. (2) The maximum of motion parameters error is 6%. (3) In the effective range, the error of distance estimate is under 3%. (4) The vision processing scheme which presented in this paper is feasible.
(keywords：AUVs, biomimetic, underwater robot vision, optical flow, camera calibration, stereo imaging, motion estimated)

中文摘要I
英文摘要II
目錄III
圖目錄V
表目錄VII
符號說明VIII
第一章 前言1
1.1 研究動機1
1.2 相關研究3
1.3 研究目的4
1.4 內容說明4
第二章 光流計算5
2.1 基本原理5
2.1.1 光流與運動場5
2.1.2 物體運動與光流之關係6
2.2 光流計算法9
2.2.1 梯度法9
2.2.2 相關法12
2.2.3 頻域法14
第三章 攝影機校正及立體成像法15
3.1 攝影機校正15
3.1.1 考慮攝影機物理模型之校正法16
3.1.2 不考慮攝影機物理模型之校正法18
3.2 立體成像法20
3.2.1 攝影機轉換關係20
3.2.2 簡化模型22
3.2.3 對應點之求取23
第四章 運動估測25
4.1 運動量計算25
4.1.1 基本假設25
4.1.2 演算法推導27
4.1.3 迴圈疊代28
4.2 測距及實際運動估測30
第五章 系統架構與驗證33
5.1 系統架構33
5.1.1 硬體系統33
5.1.2 軟體系統36
5.2 系統驗證39
5.2.1 光流及運動量計算驗證39
5.2.2 攝影機校正與距離量測43
5.3 實用例49
第六章 結論52
參考文獻54

[1] Horn, K. P. and B. G. Schunck,“Determining optical flow”, Artificial Intelligence, vol.17, pp. 185-203, 1981
[2] Fuh, C. S. and P. Maragos, “Region-Based Optical Flow Estimation”, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, San Diego, CA, pp. 130-133, 1989
[3] Heeger, D. J., “Optical flow using spatiotemporal filter”, International Journal of Computer Vision, pp. 279-302, 1988.
[4] Wu Yu-Te, Takeo Kanade, Jeffrey Cohn and Ching-Chung Li, “Optical Flow Estimation Using Wavelet Motion Model”, Proceedings of the International Conference on Computer Vision, Bombay, pp.992-998, 1998.
[5] Sutherland, I., “Three Dimensional Data input by tablet”, Proceedings of IEEE, vol. 62, pp. 453-461, 1974
[6] Yakimovsky, Y. and R. Cunningham, “A System for Extracting 3D Measurement from a Stereo Pair of TV Cameras”, Computer Graphics Image Processing, vol. 7, PP. 195-210, 1978
[7] Tsai R. Y, “A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off the Shelf TV Cameras and Lenses”, IEEE Journal of Robotics and Automation, vol. RA-3, no. 4, pp. 323-344, 1987
[8] Martins, H. A., J. R. Birk, R. B. Kelly, “Camera Models Based on Data from Two Calibration Planes”, Computer Graphics and Image Processing, pp. 173-180, 1981
[9] Gremban, K. D. and C. E. Thorpe, “Geometric Camera Calibration Using System of Linear Equation”, Proceedings of IEEE International Conference on Robotics and Automation, pp. 562-567, 1988
[10] Remy, S., M. Dhome, N. Daucher and J. T. Lapreste, “Estimating the Radial Distortion of an Optical System; Effect on a Localization Process”, Proceedings of IEEE International Conference on Image Processing, vol. 2, pp. 997-1001, 1994
[11] Pan Li, Zhang Qian, Xiao Xinshu, Ding HaiShu, Wang Guangzhi, Zhao ping, “CCD-Based system for Three Dimensional Real Time Positioning”, Proceedings of the 20th Annual International Conference of the IEEE Engineerging in Medicine and Biology Society, vol. 20, no. 5, pp. 2503-2506, 1998
[12] Juyang Weng, Paul Coen and Nicolas Rebibo, “3D Motion Estimation, Understanding, and Prediction from Noisy Image Sequences”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. PAMI-9, no. 3, pp. 370-389, May, 1987
[13] Nicholas Ayache and Francis Lustman, “Trinocular Stereo Vision for Robotics”, IEEE Trans. on Neural Network, vol. 3, no. 1, pp. 5-13, Jan, 1992
[14] Leung, M. K. and T. S. Huang, “Estimating Three Dimension Vehicle Motion in an Outdoor Scene Using Stereo Image Sequences”, International Journal of Imaging Systems and Technology, vol. 4, pp. 80-97, 1992
[15] Tan, Y. P., S. R. Kulkarni and P. J. Ramadge, “A New Method for Camera Motion Paramater Estimation”, IEEE International Conference on Image Processing, vol. 1, pp. 406-409, 1995
[16] Mitiche, Amar, “Computational Analysis of Visual Motion”, Plenum Press, NewYouk, 1994
[17] Lucas, B. and T. Kanade,“An Iterative image registration technique with an application to stereo vision”, DARPA Image Understanding Workshop, pp. 121-130, 1981
[18] Loaiza, H, J. Triboulet, S. Lelandais, C. Barat, “Matching Segments in Stereoscopic Vision”, IEEE Instrumentation and Measurement Magazine, March, pp. 37-42, 2001
[19] Anandan, P.,“A computational framework an algorithm for the measurement of visual motion”, International Journal of Computer Vision, 2:283-310, 1989
[20] Ancona, N. and T. Poggio,“Optical flow from 1d correlation: Aplication to a simple time-to-crash detector”, Massachusetts Institute of Technology Artificial Intelligence Laboratory and Center for Biological and Computational Learning, 1993
[21] Otero, J., A. Otero, R. Muniz and L. Sanchez,“Robust optical flow estimation”, Information Intelligence and Systems, 1999. Proceedings. 1999 International Conference, pp. 370-374, 1999
[22] Tekalp, A.,“Digital Video Processing” , Prentice Hall PTR,1995.
[23] Chun-Jen, T., N. P. Galatsanos and A. K. Katsaggelos,“Optical flow estimation from noisy data using differential techniques”, Acoustics Speech and Signal Processing, 1999. Proceedings, 1999 IEEE International Conference on Volume: 6, pp. 3393-3396, 1999
[24] Barron, J. L., D. J. Fleet, S. S. Beauchemin and T. A. Burkitt, “Performance of optical flow Techniques”, Computer Vision and Pattern Recognition, IEEE Computer Society Conference, pp. 236~242, 1992
[25] Simoncelli, E. P., E. H. Adelson and D. J. Heeger, “Probability Distribution of Optical Flow”, Computer Vision and Pattern Recognition, IEEE Computer Society Conference on, pp. 310~315 1991]
[26] Elad. M., P. Teo and H. O. Yacov, “Optimal Filters for Gradient-Based Motion Estimation”, Electrical and ELectronic Engineers in Israel, The 21st IEEE Convention of the, pp. 195~197, 2000
[27] Ghosal, S. and R. Mehrotra, “Zernike Moment-Based Feature Detectors”, Image Processing, Proceedings. ICIP-94., IEEE International Conference , Volume: 1, pp. 934~938 1994
[28] Fernado A. M., “A New Motion Parameter Estimation Algorithm Based on the Continuous Wavelet Transform”, Image Processing, IEEE Transactions on, Volume: 9 Issue: 5, pp. 873~888 May 2000
[29] Tseng G. and A. K. Sood, “Analysis of Image Sequences to Determine Rotational and Translational Parameters”, Intelligent Control, Proceedings., IEEE International Symposium on, pp. 174~179 1989
[30] Yi, J. W., T. S. Yang and J. H. OH, “Estimation of Depth and 3D Motion Parameters of Moving Objects with Multiple Stereo Images by Using Kalman Filter”, Industrial Electronics, Control, and Instrumentation, Proceedings of the 1995 IEEE IECON 21st International Conference on, Volume: 2, pp 1225~1230 1995
[31] Wang, D. and L. Wang, “Global Motion Parameters Estimation Using a Fast and Robust Algorithm”, Circuits and Systems for Video Technology, IEEE Transactions on , Volume: 7 Issue: 5, pp. 823~826 Oct. 1997
[32] Rodrigues, M. A. and Y. Liu, “Motion Parameter Constraints Analysis From a Single image”, Image Processing, ICIP 99. Proceedings. 1999 International Conference on , Volume: 3, pp. 704~708 1999
[33] Diamantaras, K. I., Th. Papadimitriou, M. G. Strinzis adn M. Roumeliotis, “Total Least Squares 3-D Motion Estimation”, Image Processing, ICIP 98. Proceedings. 1998 International Conference on , Volume: 1, pp. 923~927 1998
[34] Butler, D. A. and P. K. Pierson, “A Distortion-Correction Scheme for Industrial Machine-Vision Applications”, Robotics and Automation, IEEE Transactions on , Volume: 7 Issue: 4, pp. 546~551 Aug. 1991
[35] Gremban, K. D., C. E. Thorpe and T. Kanade, “Geometric Camera Calibration using Systems of Linear Equations”, Robotics and Automation, Proceedings., 1988 IEEE International Conference on , pp. 562~567 1988
[36] Jaffe, J. S., “Computer Modeling and the Design of Optimal Underwater Imaging Systems”, Oceanic Engineering, IEEE Journal of , Volume: 15 Issue: 2, pp. 101, 111 April 1990
[37] Grosky, W. I. and L. A. Tamburino, “A Unified Approach to the Linear Camera Calibration Problem”, Pattern Analysis and Machine Intelligence, IEEE Transactions on , Volume: 12 Issue: 7, pp. 663~671 July 1990
[38] Jain, R., R. Kasturi and B. G. Schunck, “Machine Vision”, McGRAW-Hill International Edtions, 1995
[39] Tekalp, A. M., “Digital Video Procession”, Prentice Hall PTR Upper Saddle River, NJ, 1995
[40] Hussain, Z., “Digital Image Processing — practical applications of parallel processing techniques”, Ellis Horwood Limited, 1991