A New Moving Object Detection Approach with Adaptive Double Thresholds

In order to improve the rate of vehicle detection, this paper proposed an adaptive double thresholds motion object mask algorithm. This novel method used multiple-frame average algorithm to initialize the background, and dynamically updated two of high and low thresholds by functional link neural network method. Meanwhile, the motion mask algorithm was used to identify the region of foreground and background to update the current background. The foreground object was extracted from dynamic double thresholds background difference method. Then combined with the mathematical morphology, the binary images became much smoother. The experimental results demonstrated that this detecting algorithm was more accurate and robust.