An improved video target tracking algorithm based on particle filter and mean-shift

Abstract

A novel tracking algorithm using particle filter algorithm combined with Mean-Shift method is proposed in this paper. To improve the real-time performance, integral histogram is merged into the particle filter tracking framework during building the target template. A Mean-Shift algorithm based on Gabor amplitude spectrum is presented to reduce the influence of similar background interference on the tracking results. Experimental results show that the proposed algorithm is of effectiveness. © 2013 Springer-Verlag.