Variable resolution image compression based on a model of visual attention

Abstract

Image compression is one of important processing techniques in visual salient object detecting. This paper proposes a model of image compression including visual attention and encoding of image. For a given image, salient regions are detected by visual attention; salient sizes are acquired by calculating feature difference between known location and periphery regions. In order to achieving variable resolution image compression, various compression ratios are adopted according to the saliency order from high to low. Original resolution is retained in the first salient region; the lowest resolution is applied in the inapparent salient regions and the middle resolution is decided by the saliency order from high to low. By this method, we achieve variable resolution image compression by the model of visual attention. Experimental results indicate that the model of image compression not only can achieve high compression ratio in total image but also can keep high resolution in salient regions and the quality of compressed image decreases as compression ratio increases. This method presents a new model for image compression.