Improved Gaussian mixture PHD for close multi-target tracking
Huanqing Zhang, Hongwei Ge, Jinlong Yang · 2014
Probability hypothesis density (PHD) filter is an optimal Bayesian multi-target filter based on random finite set. Gaussian mixture is an approximation scheme of PHD filter, which is suitable for linear Gaussian case. In multi-target tracking, when targets are moving closely to each other, GM-PHD filter cannot correctly estimate the number of targets and target states in complex tracking environment. To solve the problem, we propose an improved algorithm in this paper. The improved algorithm uses a novel method for the redistribution of target weights. The simulation results demonstrate that the proposed approach can achieve better performance compared to the other existing methods.