Moving Object Detection Based on Adaptive Loci Frame Difference Method

Criminal investigation videos often have the characteristics of complex video background, long distances of moving targets, and fast or slow-motion speeds, which have a great impact on intelligent target detection. Traditional moving target detection methods are more suitable for video environment with single background and fast-moving target. It is not sensitive and difficult to detect slow moving targets in complex backgrounds. Therefore, a moving object detection method based on improved inter-frame difference algorithm is proposed in this paper. In the process of reading each frame of video, the appropriate number of frames is determined by combining the characteristics of the image, and the difference operation is performed. Each pixel of the difference image is classified and processed. After detecting the target, the ambient noise and other interference are filtered out by setting the threshold method to realize binarization. The target region with the most effective target points is detected automatically, and the accuracy and feasibility of the algorithm are confirmed by a large number of experiments.