Improved Stereo Matching Algorithm Based on Image Segmentation

In order to solve the problem of fuzzy boundary and low texture regions prone to mismatching,this paper studied stereo matching algorithm based on image segmentation.The first step of the algorithm employs the meanshift algorithm to segment the reference image.Then,it's followed by the use of Adaptive Support Weighted SelfAdaptation dissimilarity algorithm (ASW-SelfAd) for the estimation of initial disparity.The third step is the application of Singular Value Decomposition (SVD) in solving the robust disparity plane fitting.Lastly,we apply improved clustering algorithm to merge the neighboring segments,and refine the disparity map by the new energy function.Matching experiments show that this method can not only improve the matching speed and accuracy,but also be an exact match in textureless regions,disparity discontinuous boundaries and occluded portions.