Details Preserving Deep Collaborative Filtering-Based Method for Image Denoising

TL;DR

This paper introduces Deep-CoFiB, a deep collaborative filtering method for image denoising that better preserves image details while effectively removing noise, outperforming many existing algorithms in both quantitative and qualitative assessments.

Contribution

The paper presents a novel deep collaborative filtering algorithm that enhances image denoising by balancing noise removal and detail preservation, addressing the smooth-out effect of previous neural network methods.

Findings

Deep-CoFiB achieves higher PSNR and SSIM scores.

The method preserves image details better than state-of-the-art algorithms.

It offers a fast and effective trade-off between noise reduction and detail retention.

Abstract

In spite of the improvements achieved by the several denoising algorithms over the years, many of them still fail at preserving the fine details of the image after denoising. This is as a result of the smooth-out effect they have on the images. Most neural network-based algorithms have achieved better quantitative performance than the classical denoising algorithms. However, they also suffer from qualitative (visual) performance as a result of the smooth-out effect. In this paper, we propose an algorithm to address this shortcoming. We propose a deep collaborative filtering-based (Deep-CoFiB) algorithm for image denoising. This algorithm performs collaborative denoising of image patches in the sparse domain using a set of optimized neural network models. This results in a fast algorithm that is able to excellently obtain a trade-off between noise removal and details preservation. Extensive experiments show that the DeepCoFiB performed quantitatively (in terms of PSNR and SSIM) and qualitatively (visually) better than many of the state-of-the-art denoising algorithms.