A NOVEL VISUAL TRACKING ALGORITHM BASED ON MULTI-CUES FUSION AND PARTICLE FILTER

A novel robust visual tracking algorithm, based on multi-cues fusion under a particle filter framework, is proposed in this paper. Considering the tracking accuracy and the computational simplicity, the weighed color and edges cues of the object are applied to describe the moving object. Under the particle filter framework, an adjustable measure model is incorporated into particle filter to set up the visual tracking approach. It utilizes the properties of particle filter for coping with non-linear, non-Gaussian assumption and the ability to predict the position of the moving object in a cluttered environment, meanwhile the advantages of two types of information for distinguishing moving object from the background are employed to estimate the matching likelihood ratios dynamically, according to the likelihood ratio factors, tuning the weight values of the color and edge on-line adaptively to reconfigure the optimal measure model, which ensured attaining the maximum likelihood ratio in the tracking scenario even if in the situations where the object is occluded or illumination and scale are time-variant. The indoor experimental result shows that the algorithm proposed in this paper can track the moving object accurately while the reliability of tracking in a challenging case is validated in the experimentation