Fluency DA functions approximation-based adaptive coding method of contours and regions for animation images

摘要

Our group aims to develop the unified description format of multimedia based on the Fluency information theory. In this paper, we propose the adaptive coding method of animation images as the first step to develop the unified description format of multimedia. In the proposed method, an input image is divided into multiple regions, and the contours of each region and the colors of each region are approximated adaptively based on the theory. Experimental results show that the proposed method reproduces image contours sharply and the data size of the proposed method is 83% on average compared with that of MPEG-2 coding method.