A Target Tracking Algorithm of Smart Munition Based on Multiple Features

For the accurate target tracking of smart munitions' under complex background, a improved Mean Shift tracking algorithm based on color features and edge feature is proposed to improve the robustness of the Mean Shift algorithm. Due to a color-difference space is introduced to represent color features and edge information is introduced to represent texture feature, the robustness of Mean Shift algorithm is significantly enhanced. Using color-difference space instead of hue space and using back-projection method to resolve weighted coefficients, the complexity of algorithm is reduced and the real-time demand is meet. Experimental results show that compared with the state-art Mean Shift algorithm, the proposed multiple feature Mean Shift algorithm has improved the tracking performance and suited to engineering implementation. autonomous landing.