Edge and texture aware image denoising using median noise residue U-net with hand-crafted features

Image denoising is a complex task that always yields an approximated version of the clean image. Unfortunately, the existing works have focussed only on the peak signal to noise ratio (PSNR) metric and have shown no attention to edge features in a reconstructed image. Although fully convolution neur...

Full description

Saved in:
Bibliographic Details
Main Authors: Soniya S., Sriharipriya K. C.
Format: Article
Language:English
Published: PeerJ Inc. 2025-01-01
Series:PeerJ Computer Science
Subjects:
Online Access:https://peerj.com/articles/cs-2449.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832595647264456704
author Soniya S.
Sriharipriya K. C.
author_facet Soniya S.
Sriharipriya K. C.
author_sort Soniya S.
collection DOAJ
description Image denoising is a complex task that always yields an approximated version of the clean image. Unfortunately, the existing works have focussed only on the peak signal to noise ratio (PSNR) metric and have shown no attention to edge features in a reconstructed image. Although fully convolution neural networks (CNN) are capable of removing the noise using kernel filters and automatic extraction of features, it has failed to reconstruct the images for higher values of noise standard deviation. Additionally, deep learning models require a huge database to learn better from the inputs. This, in turn, increases the computational complexity and memory requirement. Therefore, we propose the Median Noise Residue U-Net (MNRU-Net) with a limited training database without involving image augmentation. In the proposed work, the learning capability of the traditional U-Net model was increased by adding hand-crafted features in the input layers of the U-Net. Further, an approximate version of the noise estimated from the median filter and the gradient information of the image were used to improve the performance of U-Net. Later, the performance of MNRU-Net was evaluated based on PSNR, structural similarity, and figure of merit for different noise standard deviations of 15, 25, and 50 respectively. It is witnessed that the results gained from the suggested work are better than the results yielded by complex denoising models such as the robust deformed denoising CNN (RDDCNN). This work emphasizes that the skip connections along with the hand-crafted features could improve the performance at higher noise levels by using this simple architecture. In addition, the model was found to be less expensive, with low computational complexity.
format Article
id doaj-art-d7c1599a39bc423bbca878a517e0cd9c
institution Kabale University
issn 2376-5992
language English
publishDate 2025-01-01
publisher PeerJ Inc.
record_format Article
series PeerJ Computer Science
spelling doaj-art-d7c1599a39bc423bbca878a517e0cd9c2025-01-18T15:05:05ZengPeerJ Inc.PeerJ Computer Science2376-59922025-01-0111e244910.7717/peerj-cs.2449Edge and texture aware image denoising using median noise residue U-net with hand-crafted featuresSoniya S.Sriharipriya K. C.Image denoising is a complex task that always yields an approximated version of the clean image. Unfortunately, the existing works have focussed only on the peak signal to noise ratio (PSNR) metric and have shown no attention to edge features in a reconstructed image. Although fully convolution neural networks (CNN) are capable of removing the noise using kernel filters and automatic extraction of features, it has failed to reconstruct the images for higher values of noise standard deviation. Additionally, deep learning models require a huge database to learn better from the inputs. This, in turn, increases the computational complexity and memory requirement. Therefore, we propose the Median Noise Residue U-Net (MNRU-Net) with a limited training database without involving image augmentation. In the proposed work, the learning capability of the traditional U-Net model was increased by adding hand-crafted features in the input layers of the U-Net. Further, an approximate version of the noise estimated from the median filter and the gradient information of the image were used to improve the performance of U-Net. Later, the performance of MNRU-Net was evaluated based on PSNR, structural similarity, and figure of merit for different noise standard deviations of 15, 25, and 50 respectively. It is witnessed that the results gained from the suggested work are better than the results yielded by complex denoising models such as the robust deformed denoising CNN (RDDCNN). This work emphasizes that the skip connections along with the hand-crafted features could improve the performance at higher noise levels by using this simple architecture. In addition, the model was found to be less expensive, with low computational complexity.https://peerj.com/articles/cs-2449.pdfHand-crafted featuresGradient informationSkip connectionsMNRU-Net
spellingShingle Soniya S.
Sriharipriya K. C.
Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
PeerJ Computer Science
Hand-crafted features
Gradient information
Skip connections
MNRU-Net
title Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
title_full Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
title_fullStr Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
title_full_unstemmed Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
title_short Edge and texture aware image denoising using median noise residue U-net with hand-crafted features
title_sort edge and texture aware image denoising using median noise residue u net with hand crafted features
topic Hand-crafted features
Gradient information
Skip connections
MNRU-Net
url https://peerj.com/articles/cs-2449.pdf
work_keys_str_mv AT soniyas edgeandtextureawareimagedenoisingusingmediannoiseresidueunetwithhandcraftedfeatures
AT sriharipriyakc edgeandtextureawareimagedenoisingusingmediannoiseresidueunetwithhandcraftedfeatures