A Novel Vehicle Tracking and Recognition System Based on Edge Line Segment Set Matching

Image line segment feature is sparse and directional, and can facilitate the analysis of some specific geometric features and patterns. A novel vehicle target motion estimation and recognition method based on target edge line segment feature is introduced. This method includes a series of algorithms using segments as the basic elements of calculation. At first the definition of the similarities of the segment sets is presented. Then the motion estimation and target recognition algorithms are introduced. The key idea of the method is to use the similarity of segment sets as the measurement of target matching. Experimental results show significant advantages of this method, including improved efficiency, less memory requirement and better adaptiveness, over some traditional pixel-value-based algorithms in some situations.