STEREO MATCHING USING HIERARCHICAL FEATURES FOR ROBOTIC APPLICATIONS

In this paper, an efficient matching method for solving hierarchically the correspondence problem in stereo vision for robotic applications is proposed. Hierarchical features such as line segments, corners, con nectivities and regions are used for matching process, and the matchin g results at the higher level are used to guide and constrain the matc hing at the lower level. The proposed hierarchical approach makes it p ossible to maintain a globally consistent matching of features and thu s to avoid local mismatches. Experimental results show that the propos ed hierarchical algorithm can be effectively used for stereo matching.