Image saliency detection based on histogram

This paper presents a new algorithm to solve the problem of saliency detection based on histogram. Firstly, we consider the images in RGB color space and Lab color space. Secondly, histogram features are computed in six color channels, respectively. The difference between the pixel and the other pixels is defined as the feature. At the same time, 2D entropy is used to evaluate the distribution of the pixels with the same color value and which is used to adjust the weights of the pixel to the saliency map. At last, we fuse the six saliency maps to get the final map. Compared with previous methods, our saliency measure is more effective and robust as demonstrated by the experiments.