本文以嵌入式系統(EBC 320 Embedded System)為平台，以攝影機擷取2D影像，並利用NI控制卡控制伺服馬達搭配雷射尋標器(Laser Range Finder)對環境掃描來重構視覺影像。利用此兩不同設備，將視覺影像與攝影機所擷取的影像進行影像對正(image registration)。
影像對正主要分為『手動影像對正』與『自動影像對正』，首先是利用MATLAB程式語言將攝影機所擷取的影像圖片和雷射尋標器所掃描的視覺影像進行手動影像對正並做小波影像融合，目的是驗證影像對正是否可行，接著，在視窗程式開發環境下，進行自動化對正探討。利用了自動化影像對正的方法和定義視區範圍來當作輪型機器人閃避障礙物的依據。因此，實作輪型機器人穿越不同狹縫寬度障礙物與不同高度障礙物之測試。

In this thesis, an embedded system (EBC 320 Embedded System) as a platform to capture 2D camera image, and take advantage of NI control card to control servo motor with a laser range finder to scan the environment for reconstruction of visual image. Using the two different devices, visual images and images captured by camera for image registration.

The image registration is carried out with manual and automated image registration. The first is the use of MATLAB programming language, with image captured by camera and visual image scanned by laser range finder to carry out image registration manually and image fusion with wavelet transformation, the purpose is to verify the feasibility of image registration, and then, in the window program development environment for automated image registration to explore. On the basis of use of automated image registration and definitions as the visual area for wheel-robot is to avoid obstacles. These are tests for wheel-robot through obstacles in different width of slit and different heights to further validate its efficiency.

[1] Peter Shirley, Fundamental of Computer Graphics 2nd ed., pp.56-57, A K Peters, Ltd, April 2005.
[2] C. Harris and M. Stephens.“A combined corner and edge detector.”Proceedings of the 4th Alvey Vision Conference: pp. 147–151, 1988.
[3] Ardeshir Goshtasby, “Image registration by local approximation methods”, Image Vision Computing. 6, pp. 255–261, 1988.
[4] 王愛玲,葉明生,鄧秋香，MATLAB R2007圖像處理技術與應用，電子工業出版社，西元2008年1月。
[5] http://en.wikipedia.org/wiki/PSNR
[6] NEXCOM, Embedded Board Computer EBC 320 Series User’s Manual Version 1.1,August ,2007.
[7] Hirohiko Kawata, Akihisa Ohya and Shin’ichi Yuta, Wagle Santosh and Toshihiro Mori,“Development of ultra-small lightweight optical range sensor system,” IEEE/RSJ International Conference on Intelligent Robots and Systems, pp.1078-1083,August,2005.
[8] B.S. Manjunath, C. Shekhar, R. Chellapa, “A new approach to image
feature detection with applications, ”Pattern Recognition 29, pp 627–640, 1996.
[9] Q. Zheng, R. Chellapa, “A computational vision approach to image
registration, ” IEEE Transactions on Image Processing 2, pp 311–325, 1993.
[10] A. Noble, “Finding corners,” Image and Vision Computing 6, pp 121–128, 1988.
[11] W.K. Pratt, Digital Image Processing, 2nd ed., Wiley, New York,
1991.
[12] B. Zitova and J. Flusser, “ Image Registration methods: A
Survey,” Image Vision Computing, vol. 21, no. 11, pp. 977-1000, Apr., 2003.
[13] R.N. Bracewell, The Fourier Transform and Its Applications,
McGraw-Hill, New York, 1965.
[14] ActiveMedia Robotics, Pioneer 3 Operations Manual, 2005.
[15] http://www.intel.com/technology/computing/opencv/
[16] http://robots.mobilerobots.com/ARIA
[17] http://www.servocity.com/html/hs-422_super_sport_.html
[18]David Brackeen/著，范濬綸、尤宗翰/譯，大師談java遊戲設計，上奇科技出版社，2003年12月16日。
[19]郭雷,李暉暉,鮑永生，圖像融合，電子工業出版社，西元2008年2月。