Color Stereoscopic Steps for Road Obstacle Detection

In this article, we present a fast and self-adaptive stereo vision system for road obstacle detection. Both calibration and processing are detailed. Concerning color calibration, it is verified and adjusted by changing chrominance and brightness of cameras. Steps of processing are as follows: first, vertical edges points are extracted using the operator color declivity. Then, vertical edges points are associated using dynamic programming based on geometric, non-reversal, uniqueness and color photometric constraints. Finally, 3D edges of obstacle are extracted and obstacles are bounded by boxes. Performance of segmentation, matching and obstacle detection are discussed. Experimental results are shown