A Detect Motion Algorithm with Efficiency Priority for Embedded Systems

In a real-time video with surveillance embedded system, it is a hot research on an effective motion detection algorithm. In this paper, we bring forward a detected motion algorithm which is based on a 16 channels video card alarming system. This algorithm enhances the effectiveness obviously without losing the precision. By using the background shadowing algorithm, it can decrease much computation time of CPU when the background changes quickly; by using the self-adaptive algorithm for selecting the threshold, it can wipe off complex computation in the complex mathematical models, so it also can reduce the burden of CPU; the algorithm adapt to many requirements in different places and different sensitiveness. The result of experiments shows that our method can reduce the burden of computers effectively and get a better threshold compared with other methods.