Efficient image compression with detecting text regions

Various image and video compression methods have been developed so far for data reduction. However, necessary information of an original image or video may be degraded if the image is highly compressed without considering the importance of each region in the image. In this paper, we propose a method to detect regions that are important for humans in an image and compress the image efficiently by changing the compression rate according to the importance. We assume that text in an image is important and propose a method for detecting important text regions with the edge information and saliency map.