隨著電腦科學的快速進步，以及近年Kinect的興起，深度影像的應用成為了近年來熱門的研究主題。如今，影像設備的價格愈來愈便宜，和價格較高的Kinect相比，透過兩台攝影機得到深度資訊的立體匹配系統，變得更加受到重視。本論文提供快速的立體匹配系統，並使用稀疏梯度點的資訊去尋找在另一影像的所有可能的相似點，接著再用最大化事後機率的方式，來決定各相似點為符合點的可能性，最後再透過鄰近點的深度資訊，達到去除雜訊和平滑化的結果。實驗的部分，使用Middlebury Stereo Datasets內提供的左影像和右影像，以及理想深度圖；透過左影像和右影像，經由本論文的系統來得到結果，並和理想深度圖去做比較，得到精確度。

With the quick advancement of computer science and the recent rise of Kinect, the application of disparity map has become a popular research topic in the past few years. Today, the prices of imaging equipment are getting lower and lower. Compared with the more expensive Kinect, the Stereo System, which acquires depth information with two cameras, becomes highly emphasized. This thesis provides a quickly Stereo system, and uses the information of sparse gradient points to find out all the possible similar points in the other image. Then, the disparity map is used to explore the possibility of each similar point being the corresponding point. Lastly, with the depth information of the neighboring points, the noises are removed and the result becomes smoother. As for the experiment, we make use of the left and right images provided by Middlebury Stereo Datasets as well as the ground truth. We process the left and right images with the above system to find the result, while comparing it with the ground truth to obtain precision.

[1] M. Bleyer, M. Gelautz,“A Layered Stereo Matching Algorithm Using Image Segmentation and Global Visibility Constraints”, ISPRS Journal of Photogrammetry and Remote Sensing, Volume 59, Issue 3, pp.128-150, 2005.
[2] S. Birchfield, C. Tomasi, ” Depth Discontinuities by Pixel-to-Pixel Stereo”, International Journal of Computer Vision”, pp.269–293,1996
[3] C. M. Bishop and N. M. Nasrabadi, Pattern Recognition and Machine Learning, Vol.1. New York: springer, 2006
[4] A. F. Bobick, and S. S. Intille. “Large Occlusion Stereo“. International Journal of Computer Vision, pp.181–200, 1999.
[5] Y. Boykov, O. Veksler, R. Zabih, “Fast Approximate Energy Minimization via Graph Cuts”, IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.1222–1239, 2001
[6] G. Bradski, A. Kaehler, Learning OpenCV, Computer Vision with the OpenCV Library, O’Reilly, 2008.
[7] P. Cignoniz, C. Montaniz, R. Scopigno, “DeWall: A Fast Divide and Conquer Delaunay Triangulation Algorithm in Ed”, Consiglio Nazionale delle Ricerche, Via S. Maria 46, 56126 Pisa, Itall, 1998
[8] P. F. Felzenszwalb and D. P. Huttenlocher. “Efficient Belief Propagation for Early Vision”, Computer Vision and Pattern Recognition, pages I, pp. 261–268, 2004.
[9] S. Forstmann, Y. Kanou, J. Ohya, S. Thuering, A. Schmitt, “Real-Time Stereo by using Dynamic Programming”, in: IEEE Conference on Computer Vision and Pattern Recognition Workshop, vol. 3, pp. 29–36, 2004
 
[10] A. Geiger, M. Roser, R. Urtasun, W. Kubinger and M.Vincze, “Efficient Large-Scale Stereo Matching”, Asian Conference on Computer Vision, Volume 6492, pp.25-38, 2010.
[11] Y. Gong and W. Xu, Machine Learning For Multimedia Content Analysis, Vol. 30. Springer, 2007.
[12] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 3rd edition, Pearson, 2007.
[13] Y. S. Heo,“Robust Stereo Matching using Adaptive Normalized Cross-Correlation“, IEEE Transactions on Pattern Analysis and Machine Intelligence, VOL. 33, NO. 4, 2011
[14] H. Hirschmueller, P. R. Innocent, J. Garibaldi, “Real-Time Correlation-Based Stereo Vision with Reduced Border Errors”, International Journal of Computer Vision, pp.229–246, 2002
[15] M. Humenberger, C. Zinner, M. Weber, W. Kubinger and M. Vincze, “A Fast Stereo Matching Algorithm Suitable for Embedded Real-Time Systems”, Computer Vision and Image Understanding, pp.1180–1202, 2010.
[16] T. Kanade, and M. Okutomi,” A Stereo Matching Algorithm with an Adaptive Window: Theory and Experiment”, In Proc. Image Understanding Workshop, pp.383–389,1990
[17] T. Kanade. “Development of a Video-Rate Stereo Machine”, In Image Understanding Workshop, pp.549–557, 1994
[18] T. Kanade, “A Stereo Machine for Video-Rate Dense Depth Mapping and Its New Applications1”, IEEE On Computer Vision and Pattern Recognition, pp.196-202, 1996.
[19] A. Klaus, M. Sormann and K. Karner,”Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure”, Pattern Recognition, International Conference on Pattern Recognition, p15-18, 2006
[20] K. Konolige, “Small Vision Systems: Hardware and Implementation”, Artificial Intelligence Center, SRI International, 1997.
[21] D. Scharstein and R. Szeliski. “Stereo Matching with Nonlinear Diffusion”, International Journal of Computer Vision, pp.155–174, 1998.
[22] D. Scharstein, and R. Szeliski, “A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms”, International Journal of Computer Vision, pp.7-42, 2002.
[23] M. Shimizu and M. Okutomi. “Precise Sub-Pixel Estimation on Area-Based Matching”, International Conference on Computer Vision, volume I, pp.90–97, 2001.
[24] R. Szeliski and R. Zabih. “An Experimental Comparison of Stereo Algorithms”, In International Workshop on Vision Algorithms, pp.1–19, 1999.
[25] J. Sun, H. Y. Shum, and N. N. Zheng. “Stereo Matching using Belief Propagation”, IEEE On Computer Vision and Pattern Recognition, vol.25, no.7, 2003.
[26] H. Tao, H. S. Sawhney, and R. Kumar. “A Global Matching Framework for Stereo Computation”, International Conference on Computer Vision, pages I, pp. 532–539, 2001.
[27] Q. Tian and M. N. Huhns.” Algorithms for Subpixel Registration”. Computer Vision, Graphics, and Image Processing, Vol.35, pp.220–233, 1986.
[28] Z. F. Wang, Z. G. Zheng, “A Region Based Stereo Matching Algorithm using Cooperative Optimization”, Computer Vision and Pattern Recognition, pp1-8, 2008
[29] R. Zabih, J. Woodfill,“Non-Parametric Local Transforms for Computing Visual Correspondence”, in: European Conference on Computer Vision, pp.151–158,1994
[30] A. Zureiki, M. Devy, and R. Chatila, “Stereo Matching and Graph Cuts“, Computer Vision and Pattern Recognition, vol.1, pp.74-81, 2004