摘要
针对传统粒子滤波算法导致的更多粒子的估算后概率分布问题,提出了一种基于Kalman滤波的改进粒子滤波算法,使得采样粒子大多集中在高概率分布区域,提高了运算效率;同时,针对在对人体上肢运动被大面积同色干扰或遭遇长时间遮挡的情况下,提出了一种聚类结果方法,从而实现了对人体上肢的跟踪。
Aimed at the particle leads to more traditional particle filter algorithm to estimate the posterior probability distribution problems, proposed an improved particle filter algorithm based on Kalman filter, which makes the sampling par ticles are mostly concentrated in the region of high probability, improve operational efficiency. At the same time, according to the motion of the human upper limb was in large area with color interference or suffered a long time in case of occlusion, proposed a method of clustering results, the realization of the human upper limb tracking.
出处
《新技术新工艺》
2015年第3期72-74,共3页
New Technology & New Process
作者简介
董英辉（1980-），男，硕士，讲师，主要从事体育训练等方面的研究。
