Stereo Matching Algorithm Based on Detecting Feature Points

Abstract

A new method for solving the stereo matching problem in the presence of large occlusion is presented. This method for stereo matching and occlusion detection is based on searching disparity point. In this paper, we suppose that a pair of epipolar-line images is a projection of a group of piece-wise straight lines on the left and right images respective. Therefore the disparity curve corresponding to a pair of epipolar-line images may be approximated by a group of piece-wise straight lines. Then the key of solving disparity curve is how to get the “characteristic points” on the group of piece-wise straight lines. Based on this view, we fetched out the conception “disparity point”, and three kinds of special disparity points are correctly corresponding to the “characteristic point”. By analyzing intensity property of a disparity point and its neighbor points, an approach which combines stepwise hypothesis-verification strategy with three constraint conditions is devised to extract the candidate disparity points from the epipolar images.