In order to solve the problem that tracking maneuvering target under complex background and large view is easy to lose and the problem of low tracking precision

propose a fast maneuvering target detection and tracking algorithm under complicated background. Frame difference algorithm is utilized to extract maneuvering target in image

and color histogram model of maneuvering target is established in initial frame

and the subsequent input image pixel values transformed into probability value under histogram distribution

according to the similarity with the target model and pixel values of each candidate region as density. Using adaptive mean shift algorithm to search for search real position of maneuvering target. In the process of tracking. Using Kalman filtering to predict target position.The experimental results show that the algorithm can accurately fastly detect and track maneuvering targets under complex background and large view.