A novel obstacle detection method using stereo vision and two-stage dynamic programming

Abstract

Stereo vision based obstacle detection is an algorithm that aims to detect and compute obstacle depth using stereo matching and disparity map. This paper presents a novel method to detect obstacles in highly textured environments using two-stage dynamic programming. The algorithm consists of several steps including pre-processing, obstacle detection, analysis of disparity map using two-stage dynamic programming (TSDP) technique and depth computation. This method works well in highly textured environments and ideal for real applications. The disparity map for the stereo images is found in the 3D correlation coefficient volume by obtaining the global 3D maximum-surface rather than simply choosing the position that gives the local maximum correlation coefficient value for each pixel. The 3D maximum-surface is obtained using two-stage dynamic programming (TSDP) technique. An adaptive thresholding is also applied for better noise and texture removal. Experimental results show the effectiveness of the proposed method.