Adaptive Color-based Histogram Particle Filter

Object tracking is a difficult and primary task in many video processing applications. Target tracking problem is an important research subject in image processing field, in the field of pattern recognition, and so on. Along with the development of modern war network and the development of information, more and more tracking technology has receives the various county" close attention. Target tracking technology based on the visual, also has its broad application prospects in military and in the civil field.The algorithm of target tracking based on particle filter is more popular among many algorithms in recent years. Particle filter is one of the powerful tools for nonlinear/non-Gaussian state estimation problem, which is also a complete theoretical framework for target tracking. The algorithm is based on Bayesian statistic models. The tracking problem is converted into the non-Gaussian and nonlinear Bayesian filtering problem. Then target tracking can be implemented by solving the filtering problem.Because of the diversity of various video processing tasks, there exists no optimum method that can perform properly for all application. Histogram-based particle filtering is one of the most successful object tracking methods. However, for dealing with visual tracking in real world condition (such as changes in illumination and pose) is still a challenging task. In the paper, we have proposed a color-based adaptive histogram particle filtering method that can update the target method. We have used the Bhattacharyya coefficients to measure the likelihood between two color histogram.Face detection and face tracking is one of the key technology in the information processing, in intelligent interactive, video sureillance, video retrieval, video conference, and so on,it has very important application value. This article will introduce the adaptive color histogram particle filter algorithm applied to the face tracking problem. Our experimental results show that the proposed method is robust against particle occlusion, rotation, scaling, object deformation, and changes in illumination and pose. It is also fast enough to be used in real-time application.