Disparity estimation method based on reliability space using color edge

Abstract

In this paper, we propose a Disparity Estimation Method based on Reliability Space using Color edge. It first makes the disparity space image by using the difference between the Stereo images. It than calculates a section of continuous minimum value of the disparity space image, and the length of the section is calculated to produce reliability space. By comparing the parallax of the reliability space, the disparity data of each pixel selects the disparity data with the highest accuracy. Moreover, the parts that make regional boundary errors are corrected by classifying the boundary of each objects using color edge. The performance of the proposed stereo matching method is verified by various experiments. © 2010 Springer-Verlag.