Image salient region detection based on histogram

Image salient region detection is useful for applications like image segmentation, image retrieval et al. According to the thought that the color of salient region is global scarcity and the spatial distribution is compact, a new image salient region detection method is proposed in this paper. The method estimates the salient map based on histogram by using the pixel color differences and the spatial distribution characteristics in both Lab and HSV color spaces separately. Then the final saliency map is built by normalizing and fusing the above saliency maps. The publicly ImgSal dataset is used for experiment evaluation, which has six different types of sub-datasets. Compared with seven other state-of-the-art methods, the proposed method produces superior performance. Experimental results show that the proposed method is effective.