卢艳军, 王诗宇, 张太宁, 赵为平. 针对复杂环境问题的改进跟踪算法研究[J]. 微电子学与计算机, 2020, 37(9): 78-82.
引用本文: 卢艳军, 王诗宇, 张太宁, 赵为平. 针对复杂环境问题的改进跟踪算法研究[J]. 微电子学与计算机, 2020, 37(9): 78-82.
LU Yan-jun, WANG Shi-yu, ZHANG Tai-ning, ZHAO Wei-ping. Research on improved tracking algorithm for complex environmental problems[J]. Microelectronics & Computer, 2020, 37(9): 78-82.
Citation: LU Yan-jun, WANG Shi-yu, ZHANG Tai-ning, ZHAO Wei-ping. Research on improved tracking algorithm for complex environmental problems[J]. Microelectronics & Computer, 2020, 37(9): 78-82.

针对复杂环境问题的改进跟踪算法研究

Research on improved tracking algorithm for complex environmental problems

  • 摘要: 为解决无人机跟踪目标时面对的复杂环境问题,选择以基于颜色特征进行跟踪的MeanShift算法为基础,对其作出改进.针对无人机跟踪目标时,目标颜色与环境背景相似的问题,提出改进MeanShift算法的跟踪特征,将相机采集的视频格式由RGB颜色空间转化为HSV颜色空间,并选取其中的色调分量作为跟踪特征;针对无人机跟踪过程中目标被短期遮挡的问题,提出将改进后的MeanShift算法与Kalman滤波相结合,利用Kalman滤波的预测功能,使算法在面对目标被短期遮挡的情况下依旧能稳定跟踪.通过实验,验证了本文提出的跟踪算法的有效性.

     

    Abstract: In order to solve the complex environmental problems faced by the drone tracking target, the MeanShift algorithm based on color feature tracking is selected and improved. Aiming at the problem that the target color is similar to the environment background when the target is tracked by the drone, the tracking feature of the MeanShift algorithm is proposed. The video format captured by the camera is converted from the RGB color space to the HSV color space, and the hue component is selected as the tracking feature. For the problem that the target is short-term occlusion during the tracking process of the UAV, the improved MeanShift algorithm is combined with Kalman filtering, and the prediction function of Kalman filtering is used to make the algorithm stable in the face of short-term occlusion. The effectiveness of the tracking algorithm proposed in this paper is verified by experiments.

     

/

返回文章
返回