Image denoising based on improved data-driven sparse representation

TL;DR

This paper introduces an improved data-driven sparse representation method for image denoising that reduces the number of filters used, leading to better recovery quality and faster computation.

Contribution

It proposes a novel sparse representation system with fewer filters, enhancing denoising performance and efficiency over existing data-driven tight frame methods.

Findings

Outperforms original method in denoising quality

Reduces computational time significantly

Achieves better noise removal with fewer filters

Abstract

Sparse representation of images under certain transform domain has been playing a fundamental role in image restoration tasks. One such representative method is the widely used wavelet tight frame systems. Instead of adopting fixed filters for constructing a tight frame to sparsely model any input image, a data-driven tight frame was proposed for the sparse representation of images, and shown to be very efficient for image denoising very recently. However, in this method the number of framelet filters used for constructing a tight frame is the same as the length of filters. In fact, through further investigation it is found that part of these filters are unnecessary and even harmful to the recovery effect due to the influence of noise. Therefore, an improved data-driven sparse representation systems constructed with much less number of filters are proposed. Numerical results on denoising experiments demonstrate that the proposed algorithm overall outperforms the original data-driven tight frame construction scheme on both the recovery quality and computational time.