论文部分内容阅读
针对无人机目标跟踪过程中CamShift算法对目标颜色相似背景干扰和遮挡干扰鲁棒性差问题,对CamShift算法进行了改进。首先,针对CamShift算法模板信息单一,易受到颜色相似背景干扰的问题,提出基于H分量和LBP二维直方图模板的CamShift目标跟踪算法,改进算法提高了算法对相似目标干扰的鲁棒性,且有效帧率提高了约21%;针对目标跟踪过程中目标易受到障碍物遮挡的问题,在CamShift算法中引进了Kalman滤波预测机制,增强了跟踪算法在目标遮挡条件下的鲁棒性和跟踪效率,其中跟踪效率提高了约25%,每帧迭代所用时间下降了约36%。
To improve the CamShift algorithm, aiming at the poor robustness of CamShift algorithm to target-color similar background interference and occlusion jamming in UAV target tracking process, CamShift algorithm is improved. Firstly, a CamShift target tracking algorithm based on H-component and LBP two-dimensional histogram template is proposed for the problem that the template information of CamShift algorithm is single and vulnerable to color-like background interference. The improved algorithm improves the robustness of the proposed algorithm to similar target interferences The effective frame rate is increased by about 21%. In view of the problem that the target is easily obstructed by obstacles in the target tracking process, Kalman filter prediction mechanism is introduced in CamShift algorithm to enhance the robustness and tracking efficiency of the tracking algorithm under the target occlusion condition , Of which tracking efficiency increased by about 25% and the time spent on each iteration decreased by about 36%.