A PROGRESSIVE CORRESPONDENCE APPROACH

The binocular stereo vision, based on D.Marr's computational stereo theory, is an efficient approach for recovering range information from 2D images, in which the stereo correspondence is its key part. A stereo matching approach based on edge feature is proposed in this paper. The procedure of the correspondence is fulfilled through the edge detection,the similarity check and the consistence check, controlled by a matching strategy relaxing gradually the constraints. At the same time, for the solution of higher false matching rate in horizontal edges, a method of disparity limitation is suggested. The experimental result shows that the matching approach proposed is efficient and feasible.