Research on Vehicle Detection Method Based on Video in Color Low Illumination Environment

In this paper, we study the method of video vehicle detection in color low illumination environment at night. At first, the low illumination image enhancement algorithm based on multi-scale Retinex used to process the original color image, and we make the open operation noise reduction processing after extracting moving object by frame difference method. Because of the extracted motion vehicle has tomography, this paper presents an effective solution to determine whether the moving target is the same vehicle. In this way we solve the problem of the faultage. The results of experiments show the proposed method in this paper has better performance. The method not only meets the requirement of real-time vehicle detection, but also solves the problem of detecting moving vehicles at night.