Image interpolation based on decomposition

Abstract

This paper proposes a new approach for interpolating natural images. Unlike other conventional interpolation methods, we exploit the characteristics of an image and decompose it into a texture part and a non-texture (cartoon) part. In the non-texture part, we interpolate it with bicubic interpolation which performs well for smooth areas. We deal the texture part with peak transform based interpolation which resets the image pixels to capture the characters of the texture through peak transform. The experiment results demonstrate our new interpolation method has higher PSNR and better subjective quality of the interpolated image, especially in the texture part.

Publication
2010 International Symposium on Intelligent Signal Processing and Communication Systems
Li Song
Li Song
Professor, IEEE Senior Member

Professor, Doctoral Supervisor, the Deputy Director of the Institute of Image Communication and Network Engineering of Shanghai Jiao Tong University, the Double-Appointed Professor of the Institute of Artificial Intelligence and the Collaborative Innovation Center of Future Media Network, the Deputy Secretary-General of the China Video User Experience Alliance and head of the standards group.