A New Robust Object Tracking Algorithm Based on Multi-Feature Fusion

The object tracking by single feature often leads to poor robustness. In this paper, an object tracking algorithm based on multi-features fusion is presented. An adaptive method of choosing object color histogram is presented and the histogram is background weighted in order to get an accurate color model of the object. At meanwhile, then spatiograms feature is applied to obtain spatial layout of these color information. These features are rationally fused in the framework of Particle filter. The uncertainty measurement method is then introduced into features fusion to adjust the relative contribution of different features adaptively. Experimental results indicate that the proposed method is robust and highlights good performance in complex scene.