Saliency detection by non-linear intensity mapping in images

Visually saliency detection provides an alternative methodology to image description in many applications such as adaptive content delivery and image retrieval. In this paper, we propose a novel computational method to model visual saliency in nature images. To efficiently eliminate high contrast noise regions in the background, we propose a constructive approach of non-linear intensity mapping for rendering an image based on local and global context information. Therefore, the most salient region can be easily selected as the one which is globally most isolated and has locally distinctive characteristics. The proposed approach intrinsically provides an filternative methodology to model attention with low implementation complexity. Experiments show that our approach achieves much better performance than that from the existing state-of-art.