A fast stereo matching method for real time vehicle front perception with linear cameras

This paper presents a passive stereo-based approach for real time vehicle front perception with linear cameras. The heavily investigated problem in stereo approaches is the matching between two or more images of a scene observed, by two or more video cameras, from different viewpoints. It consists of identifying features in the left and right images that are projections of the same physical feature in the three-dimensional space. In our method, the correspondence problem is first mapped onto a two-dimensional matrix, called matching matrix. The problem is then formulated in terms of finding true matches that satisfy local and global constraints. Local constraints are used to discard impossible matches whereas global ones are used to search best matches. The performance of the proposed stereo matching method is evaluated on stereo images acquired in real traffic conditions.