基于改进Camshift的无人机目标跟踪算法研究

Research on target tracking of UAV based on improved Camshift algorithm

无人机在对地面目标进行跟踪时,传统Camshift算法易受相似颜色背景/目标、遮挡等干扰.本文提出一种改进Camshift的目标跟踪方法.通过提取跟踪目标的色度、饱和度和LBP纹理特征分量建立基于三维联合直方图的跟踪模板,并采用自适应加权策略来调整三种特征分量的权重值,提高算法的跟踪准确度;在跟踪目标受到遮挡干扰时,引入Kalman滤波机制,增强算法的鲁棒性.实验结果表明,改进后的算法能够满足无人机对目标跟踪准确性与实时性的要求.

When UAV tracks the ground target,the classicCamshift algorithm is susceptible to similar color background/target,occlusion,etc.The target tracking method based on improved Camshift is proposed A tracking template based on three-dimensional joint histogram is established by extracting the Hue,Saturation and LBP texture features of the tracking target,and the weighting values of the three feature components are adjusted by the adaptive weighting strategy to improve the tracking accuracy of the algorithm.When the tracking target is blocked,Kalman filtering mechanism is introduced to enhance the robustness of the algorithm.The experimental results show that the improved algorithm can meet the requirements of target tracking accuracy and real-time for UAV.