The particle filter algorithm based on color feature has been a research focus in tracking of moving object.This paper presents a particle filter algorithm based on a weighted color histogram to track moving object,using Bhattacharyya distance to describe the color similarity between the particle and the target area,which improves the measurement matching and sample weight updating.The similarity between target template and particle area is calculated,which can locate the target accurately and efficiently.The implementation of this method produces robust results for different situations such as partial occlusion,rotation and shape distortion.The results show that the method is feasible and can perform well in video surveillance and vehicle tracking.