A SURVEY ON STEREO MATCHING TECHNIQUES

Abstract

Tracing of corresponding phenomena (matching) is necessary in many applications, like image sequence analysis, aerotriangulation and 3-D geometric information extraction from stereo pairs of images. An important industrial application is robot stereo vision. In digital images the correspondence problem can be approached by signal, feature or relational matching. Common, at the moment, is the signal approach. Because of its limitations more and more image structures (features) and their mutual relationships are used. The aim of this paper is to review the current stereo matching techniques.