A Training-Based Method for Reducing Ringing Artifact in BDCT-Encoded Images

Abstract

The quantization procedure of block-based discrete cosine transform (BDCT) compression (such as JPEG) introduces annoying visual artifact. In this paper, we propose a novel training-based method to reduce the ringing artifact in BDCT-encoded high-contrast images (images with large smooth color areas and strong edges/outlines). Our main focus is on the removal of ringing artifact that is seldom addressed by existing methods. In the proposed method, the contaminated image is modeled as a Markov random field (MRF). We learn the behavior of contamination by extracting massive number of artifact patterns from a training set. To organize the extracted artifact patterns, we use the tree-structured vector quantization (TSVQ). Instead of post-filtering the input contaminated image, we synthesize an artifact-reduced image. We show that substantial improvement (both statistical and visual) is achieved using the proposed method. Moreover, since our method is non-iterative, it can remove artifact within a very short period of time.