CAMShift Tracking Algorithm Based on Multi-Feature Fusion and Kalman Filter
A CAMShift target tracking algorithm based on multi-feature fusion and Kalman filtering is proposed to solve the problem of tracking failure when CAMShift algorithm is blocked by color and occlusion in practical applica-tion scenarios.Multi-feature fusion is to fuse edge,texture and color features together based on CAMShift algorithm.The improved Canny operator is used to describe the edge feature,and the N-LBP in the unified mode is used to con-struct the texture feature.The Bhattacharyya coefficient is used to calculate the adaptive fusion weight of each feature.Through the complementary advantages between different features,the expression ability of the feature is enhanced.When there is no occlusion,CAMShift algorithm is used to calculate the target position and update the Kalman filter parameters.When there is occlusion,Kalman filter is used to predict the current target position.The simulation results show that the algorithm is less affected by the environment and the tracking error is significantly reduced compared with CAMShift algorithm.
NETL
NSTL
万方数据
