基于联合粒子滤波和Mean-Shift的运动目标跟踪算法

A Moving Target Tracking Algorithm Based on the Particle Filter and Mean- Shift

Mean Shift跟踪算法能有效实时对运动目标进行跟踪,但是当目标运动速度过快或者目标被遮挡时,跟踪效果会下降;粒子滤波跟踪算法适用于非线性非高斯环境,其跟踪效果与粒子数息息相关,数量少了导致粒子匮乏,数量多了会影响实时性。为解决上述问题,本文将mean shift和粒子滤波融合。实验表明,本文算法在运动目标速度较快和目标被遮挡时均取得了较好的跟踪效果。

The moving target can be tracked in real time by mean shift, but when the speed of the target is too fast or the target is obscured, the tracking performance will be affected. Particle filter tracking algorithm is suitable for nonlinear and non-Gaussian environment, and its tracking effect is closely related to the number of particles; the small number results in particle scarcity and the large number results in the real-time. To solve these problems, this article will combine mean shift and particle filter. Experiments show that when the target is moving fast and the target is obscured, the algorithm has achieved good tracking results.