Preceding Vehicle Detection Based on Multi-characteristics Fusion

Abstract

Preceding vehicle detection is a major research field of safety assistant driving system. This paper proposed a preceding vehicle detection method based on multi-characteristics fusion. First, the areas of interesting for preceding vehicles are established by using the bottom shadows. Then the preceding vehicles are identified through the fusion of vehicle margin characteristics, texture characteristics and vertical edge symmetry. Finally, vehicle location can be acquired by means of vehicle margin information. Experimental results show that this method has good reliability and accuracy.