Visual saliency detection based on region contrast and guided filter

Abstract

The main challenge of previous saliency detection method is the low quality of obtained saliency map which missed the edge and texture information easily. So it cannot reflect the integrated image salient information. Considering this problem, we propose a novel saliency measure method which combine region contrast and fast guided filter. This method utilizes region contrast method to obtain initial saliency maps. Then we optimize the saliency maps by using the fast guided filter. Extensive experimental results on natural image show the effectiveness of the proposed method. One aspect, the obtained final saliency maps have obvious advantages in dealing with the texture and weakening the inconsequential region. Another aspect, evaluation on the two databases validates that our method achieves superior results and outperforms compared previous approach in both precision and recall.