An improved detection algorithm of moving vehicles based on computer vision

Moving vehicle detection based on computer vision is an important aspect in the application of Intelligent Transportation Systems (ITS). How to get the accurate parameters of the vehicle in real-time, especially in complicated background situation is very critical. In this paper an improved detection algorithm of moving vehicle is proposed. According to the characteristics of some present detection algorithms, the paper makes improvement to background subtraction and symmetric difference method respectively, and combines both of them. In the method of background difference, the selective statistics background updating algorithm is proposed. Add region-growing segmentation and the morphological filtering methods to the post-processing step. Through simulation experiment with the real traffic video image, the effect is obvious.