一种抗遮挡自适应粒子滤波目标跟踪方法

An anti-occlusion method for object tracking based on adaptive particle filter

针对传统视觉跟踪算法在目标发生遮挡时容易发生偏离或失败的缺陷,提出了一种新的抗遮挡自适应粒子滤波(PF)目标跟踪方法。在粒子传播过程中,利用目标SSD(sum of squareddifference)残差所生成的高似然区域能自适应地调整状态空间中的粒子采样区域范围和采样粒子数量,使跟踪中粒子采样覆盖目标的各种状态可能性,全面提高状态空间质量。预测状态和粒子估计状态通过噪声协方差很好地融合起来,能够较有效地解决遮挡情况下的跟踪问题,使目标定位更加精确。粒子数量的自适应不仅能很好提高跟踪精度,而且在一定程度上降低了计算代价。实验结果表明,本文算法对跟踪目标遮挡具有较好的容错性和跟踪鲁棒性,能有效实现复杂场景下的目标跟踪。

A new anti-occlusion method for object tracking is presented to solve the problem that tradi- tional visual tracking algorithm often deviates or loses the targets under occlusions. The high likelihood areas generated by the sum of squared difference （SSD） residual can adjust the range and quantity of particle sampling in the state-space. The sampling method can cover various possibilities of object state and improve the quality of the state-space exploration in the diffusion process of the particle filter （PF）. The object state of forecast and estimation fused by noise covariance can achieve reliable tracking per- formanee under occlusion and gain the optimal location of ohject. The adaptive quantity of particle sam- pling not only can improve the precision, but also can reduce the computational load in a certain extent effectively. Experimental results show that the method has strong robustness and error-tolerance to oc- clusion of tracking objects, and has good performance under complex background.