基于图像序列的人体运动跟踪算法

发布时间：2019-04-21 17:54

【摘要】：为了提高人体运动的跟踪精度,提出基于图像序列的人体运动跟踪算法。首先对当前人体跟踪算法的研究现状进行分析,指出粒子滤波算法进行人体运动跟踪的不足;然后对粒子滤波算法进行改进,增加了采样粒子多样化,提高非线性人体运动目标跟踪性能,加快人体运动跟踪速度;最后采用仿真实验对人体运动跟踪算法的性能进行测试。实验结果表明,相对于其他人体运动跟踪算法,该算法提高了人体运动跟踪的准确性,而且人体运动跟踪的时间减少,具有更好的稳定性。
[Abstract]:In order to improve the tracking accuracy of human motion, a human motion tracking algorithm based on image sequence is proposed. Firstly, the current research status of human body tracking algorithm is analyzed, and the deficiency of particle filter algorithm in human motion tracking is pointed out. Then the particle filter algorithm is improved to increase the diversity of sampling particles to improve the tracking performance of nonlinear human moving targets and speed up the human motion tracking. Finally the performance of the human motion tracking algorithm is tested by simulation experiments. The experimental results show that compared with other human motion tracking algorithms, the proposed algorithm improves the accuracy of human motion tracking, and the time of human motion tracking is reduced, so it has better stability.
【作者单位】：乐山师范学院;
【基金】：四川省科技厅支撑计划项目(2014SZ0107)
【分类号】：TP391.41;TN713

【相似文献】