The moving detection algorithm based on clustering technology

Moving detection algorithms are very popular and widely applied. However, the algorithms may have lower accuracy or higher time complexity. To improve that, a moving detection algorithm based on clustering technology is proposed in the paper. It divides an image differences into some clusters. A minimum spanning tree is generated based on those clusters. Compare the average shortest path of minimum spanning tree and a threshold to detect the differences. The algorithm inherits the high accuracy of clustering technology and shows high detection performance.