Binocular Vision Based Drivable Region Fast Detection for Indoor Mobile Robot

Abstract

A fast method based on binocular vision is proposed for mobile robot to detect drivable regions. First, the image is segmented into regions by searching contours. Second, part obstacle regions are determined by the vanishing line. Then, according to the different distribution of feature points extracted from the regions under the vanishing line, we use two different method to classify regions: various constraints-based region classification is used to classify regions including many feature points and homography-substraction-based region classification is used to classify regions including rare feature points. Finally, combining the two classification methods, we get the result of drivable region detection. The results of indoor and outdoor experiments show that the method can detect drivable regions quickly and robustly.