Search

In the video sequence, during the long-term tracking process, the target and the background condition will undergo unpredictable changes, make target tracking difficult. To improve the target occlusion and large deformation in complex scenes, this paper proposes a correlation filter tracking algorithm based on fusion feature and time regularization. Firstly, the feature dimension reduction is performed on the target to form the feature matrix, which enhances the classification ability of the filter. Secondly, the time regularization is introduced to construct the target relocation under the occlusion condition. In addition, alternating direction method is used by the alternating direction method of the multiplier, to eliminate the boundary effect. Finally, template is performed by perframe. The experimental results show that the algorithm has an accuracy of 72.9 % and it can get a success score of 61.8%. By introducing time regularization, the algorithm can successfully track the target in the presence of occlusion. At the same time, it can adapt well to large changes in appearance. The algorithm performs well in terms of accuracy, robustness and speed, and can track targets in real time.