Post-processing coding artefacts for JPEG documents

Coding artefacts, including ringing and blocking artefacts, are often introduced when document images are compressed using the JPEG standard. These artefacts severely impact visual perception of the image content. Although a number of methods have been presented to deal with coding artefacts, most of them are dedicated to natural images; few works have investigated to work on document content. The current work is an attempt to fill this lack. In contrast to all the approaches taken by previous works, we propose to post-process the coding artefacts by estimating the quantization noise, which is not available on the decoder’s side. The estimated noise is then used to reconstruct the image with better quality. A number of experiments were conducted to show the efficiency of the proposed method in comparison with the state-of-the-art methods.