Road Scene Analysis using Trinocular Stereo Vision

Abstract

Road scene analysis in 3D driving environment, which aims to detect objects from continuously changing background, is vital for driver assitance system and Adaptive Cruise Control (ACC) applications. Laser or millimeter wave radars have shown good performance in measuring relative speed and distance in highway driving environment. However the accuracy of these systems decreases in an urban traffic environment as more confusion occurs due to the factors such as parking vehicles, guardrails, poles and motorcycles. A stereovision based sensing system provides an effective supplement to radar-based road scene analysis with its much wider field of view and more accurate lateral information. This paper presents an efficient solution for road scene analysis using a trinocular stereo vision based algorithm. In this algorithm, trinocular stereo vision detects all types of objects in road scene. And “U-V-disparity" concept is employed to analyze 3D geometric feature of road scene. The proposed algorithm has been tested on real road scenes and experimental results verified its efficiency.