Enhancing sparse representation of color images by cross channel transformation

TL;DR

This paper proposes a method to improve color image compression by applying cross-channel transformations and sparse approximation techniques, resulting in better quality at high compression rates compared to JPEG and WebP.

Contribution

It introduces a novel approach combining cross-channel transformations with atomic decomposition for enhanced sparsity in color image representation.

Findings

Significant improvement in image compression quality at high rates.

Competitive results with JPEG2000 standard.

Effective use of DCT and greedy algorithms for sparse coding.

Abstract

Transformations for enhancing sparsity in the approximation of color images by 2D atomic decomposition are discussed. The sparsity is firstly considered with respect to the most significant coefficients in the wavelet decomposition of the color image. The discrete cosine transform is singled out as an effective transformation for this purpose. The enhanced feature is further exploited by approximating the transformed arrays using an effective greedy strategy with a separable highly redundant dictionary. The relevance of the achieved sparsity is illustrated by a simple encoding procedure. On a set of typical test images the compression at high quality recovery is shown to significantly improve upon JPEG and WebP formats. The results are competitive with those produced by the JPEG2000 standard.