A real-time obstacle detection algorithm for the visually impaired using binocular camera

Abstract

In this paper, a real-time depth-data based obstacle detection to assist the visually impaired people in avoiding obstacles independently is presented. Depth data obtained by the binocular camera, is analyzed to detect obstacles. With the help of the proposed method, the distance between the binocular camera and the obstacle can be calculated with the speed of 30fps. Our method further allows the computation of the position and the size of the obstacle. The proposed algorithm has been extensively tested on both real images and public Laundry data-set. Experimental results demonstrate that the proposed method is not only able to detect the obstacles correctly but it is also fast, efficient and stable.