Target tracking is to search the most similar parts to target in video sequences，which is an important research in computer vision．Focus on the moving target tracking in video sequences, a method based on meanshift for target tracking is proposed. First, the target area is determined and the target model is constructed in the tracking process. Then, the candidate target model in the Nth frame is calculated. Based on the Bhattacharyya similarity function, meanshift algorithm searches iteratively for the beat candidate area of the target model in the subsequent frames. Finally, the target tracking system based on meanshift algorithm is built with MATLAB, and through a series of experiments show that this method has a good tracking performance, and has low time complexity, which can meet the need of realtime processing.