# A Dual Sparse Decomposition Method for Image Denoising

## TL;DR

This paper introduces a dual sparse decomposition technique for image denoising, especially effective under strong noise conditions, by decomposing dictionaries based on atom occurrence frequency, leading to superior denoising performance.

## Contribution

It presents a novel dual sparse decomposition approach that enhances image denoising by utilizing a new criterion for sub-dictionary selection based on atom frequency.

## Key findings

- Outperforms state-of-the-art denoising methods in PSNR and SSIM.
- Improves subjective visual quality of denoised images.
- Effective under strong noise conditions.

## Abstract

This article addresses the image denoising problem in the situations of strong noise. We propose a dual sparse decomposition method. This method makes a sub-dictionary decomposition on the over-complete dictionary in the sparse decomposition. The sub-dictionary decomposition makes use of a novel criterion based on the occurrence frequency of atoms of the over-complete dictionary over the data set. The experimental results demonstrate that the dual-sparse-decomposition method surpasses state-of-art denoising performance in terms of both peak-signal-to-noise ratio and structural-similarity-index-metric, and also at subjective visual quality.