关键词: 阴影分割, 目标提取, 交通监控

Abstract: Vehicle shadow segmentation is one of the most important parts of vehicle detection in the field of intelligent traffic monitoring. Shadow segmentation directly influences the veracity of vehicle recognition even the performance of the whole monitoring system. To cover the shortages of RGB or HSV color space based vehicle shadow segmentation algorithms, this paper puts forward a new shadow segmentation algorithm based on YCbCr color space. First, the motion area which includes the vehicle and the shadow is selected, and then the original data of the shadow according to the characteristics of the occurrence of shadow is chosen, finally, the shape and location of the shadow region is determined by the YCbCr shadow segmentation algorithm. Actual road test shows that the integral vehicle shadow can be obtained by the proposed algorithm. The algorithm with better robustness has a practical value in the field of intelligent traffic monitoring.

Key words: shadow segmentation, object extraction, traffic monitoring