Target tracking algorithm based on dynamic template and Kalman filter

The goal of target tracking is to find the targets between the consecutive frames in image sequences. Many tracking algorithms have been proposed and implemented to overcome difficulties that arise from noise, occlusion, clutter, and changes in the foreground objects or in the background environment. For the tracking methods based on traditional Kalman filter algorithm, there are several candidate areas in detection process because of the background noises in images, which lead to the wrong track results or missing the target in a video sequence. We propose a target tracking algorithm that combines Kalman filter with the dynamic template. The dynamic template is the selection criteria of target model in the next frame; it can select the best candidate region as target model by calculating the distance between the dynamic template and candidate areas. The experiments show the algorithm has better tracking precision and accuracy, and it has good robustness to background interference.