A Motion Vehicle Detection Method Based on Self-Adaptive Background Subtraction with Cumulative Inter-Frame Difference
Hong Yuan Zheng, Wen Ju An, Zhen Li · Advanced materials research · 2013
Against the poor accuracy of the vehicle counters extracted by existing vehicle detection technology, a motion vehicle detection method based on self-adaptive background subtraction with cumulative inter-frame difference is proposed in this paper. Cumulative inter-frame difference is used to subtract binary object mask. According to the binary object mask, in the area of moving objects the pixels of last background are used to modify the current background, otherwise the pixels of current image are used. The result of this operation is the current background. Then the background difference method is used to detect moving vehicles.