A Correlation Particle Filter Target Tracking Algorithm Based on Adaptive Feature Fusion

In order to improve the tracking accuracy and success rate of correlation filtering algorithm in the case of severe occlusion and background interference, a correlation particle filter target tracking algorithm based on adaptive feature fusion is proposed. The adaptive fusion of directional gradient histogram features and convolution features improves the expression ability of target features. The correlation filter and particle filter are fused to form the correlation particle filter tracker, which enhances the performance of the tracker. The secondary positioning strategy is introduced to quickly locate the target when judging the failure of target tracking. The algorithm is tested with OTB-100 video sequence and compared with the other two tracking algorithms. The experimental results show that the proposed algorithm performs well in the case of serious occlusion and background interference.