A Target Tracking Algorithm Based on Multi-Feature Fusion

Abstract

In order to improve the tracking accuracy of fast discriminative scale space tracking (fDSST) algorithm when the target is seriously occluded and rotated, this paper proposes a target tracking algorithm based on multi-feature fusion. In this algorithm, the fusion features are used to obtain more feature information of the target, and a high confidence strategy is introduced to reduce the probability of model drift when the target is obscured. OTB video sequence is used to test the algorithm, and compared with the other two tracking algorithms. The experimental results show that the algorithm proposed in this paper performs well when the target is seriously obscured and rotated.