基于Mean Shift和粒子滤波的运动目标跟踪

Moving target tracking based on Mean Shift and Particle Filter

针对Mean Shift算法跟踪效果不佳以及粒子滤波算法计算量大且实时性不强等问题,提出了一种结合Mean Shift和粒子滤波的运动目标跟踪融合算法。首先用MeanShift算法进行跟踪,在跟踪结果不佳的情况下用粒子滤波算法进行修正。实验结果表明,融合算法很好地结合了两种算法的优点,既保留了Mean Shift算法的实时性,又很好地体现了粒子滤波算法的鲁棒性,实用性很强。

The effect of Mean Shift tracking algorithm is not good and particle filter algorithm has a large amount of calculation and its real time is also not good. This paper puts forward a movement target tracking fusion algorithm combined with Mean shift and particle filter. At first, tracking with the Mean Shift algorithm, if the tracking result is not good, then using the particle filter algorithm to fix. The experimental results show that the fusion algorithm combines the two algorithm advantages well, keeps Mean Shift algorithm of real-time, and embodies the particle filter algorithm＇s robust well, the usability is very strong.