Research on image matching similarity criterion based on maximum posterior probability

Image matching is an important part for visual tracking. The Bhattacharyya coefficient is an efficient method in image statistical feature matching. But for the influence of background feature, the optimal location obtained by Bhattacharyya coefficient may not be the exact target location. Thus, biased or even wrong location may be got in visual tracking. This paper presents an image matching similarity criterion based on maximum posterior probability. The new criterion applies the statistical feature of the searching region, effectively reduces the influence of background feature, and emphasizes the importance of target feature, which distinctly improves the peak modality of matching function compared to that of Bhattacharyya coefficient. The computation complexity of the new criterion is relatively low, and the global optimal solution can be easily obtained. Compared with the matching criterion of Bhattacharyya coefficient, experimental results demonstrate that the proposed matching criterion has stronger object detection ability in complex background.