According to the disadvantage of large computation of stereo matching algorithm based on graph cuts,a new stereo matching algorithm is presented.By changing adaptively the window based on image edge features,the SSD matching method is used as a similarity decision criterion to determine the initial disparity map.Then,a left-right conformity verification is adopted to remove mistake matching points.when the energy function is constructed,the initial disparity is used as a reference of function to get the optimization of disparity distribution of the minimum global energy by graph cuts.The proposed algorithm is tested with standard stereo images and compared with other methods,experimental results show that proposed algorithm not only has better performance for processing the large low texture areas and the occluded pixels,but also has shorter matching time,which is two-thirds shorter than that of original algorithm.