Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement

Abstract Low-light image enhancement aims to enhance the visibility and contrast of low-light images while eliminating complex degradation issues such as noise, artifacts, and color distortions. Most existing low-light image enhancement methods either focus on quality while neglecting computational...

Full description

Saved in:
Bibliographic Details
Main Authors: Junfeng Wang, Shenghui Huang, Zhanqiang Huo, Shan Zhao, Yingxu Qiao
Format: Article
Language:English
Published: Nature Portfolio 2024-11-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-024-81706-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850216137094070272
author Junfeng Wang
Shenghui Huang
Zhanqiang Huo
Shan Zhao
Yingxu Qiao
author_facet Junfeng Wang
Shenghui Huang
Zhanqiang Huo
Shan Zhao
Yingxu Qiao
author_sort Junfeng Wang
collection DOAJ
description Abstract Low-light image enhancement aims to enhance the visibility and contrast of low-light images while eliminating complex degradation issues such as noise, artifacts, and color distortions. Most existing low-light image enhancement methods either focus on quality while neglecting computational efficiency or have limited learning and generalization capabilities. To address these issues, we propose a Bilateral Enhancement Network with signal-to-noise ratio fusion, called BiEnNet, for lightweight and generalizable low-light image enhancement. Specifically, we design a lightweight Bilateral enhancement module with SNR (Signal-to-Noise Ratio) Fusion (BSF), which serves the SNR map of the input low-light image as the interpolation weights to dynamically fuse global brightness features and local detail features extracted from a bilateral network and achieve differentiated enhancement across different regions. To improve the network’s generalization ability, we propose a Luminance Normalization (LNM) module for preprocessing and a Dual-Exposure Processing (DEP) module for post-processing. LNM divides the channels of input features into luminance-related channels and luminance independent channels, and reduces the inconsistency of the degradation distribution of input low-light images by only normalizing the luminance-related channels. DEP learns overexposure and underexposure corrections simultaneously by employing the ReLU activation function, inverting operation, and residual network, which can improve the robustness of enhancement effects under different exposure conditions while reducing network parameters. Experiments on the LOL-V1 dataset shows BiEnNet significantly increased PSNR by 8.6 $$\%$$ and SSIM by 3.6 $$\%$$ compared to FLW-Net, reduced parameters by 98.78 $$\%$$ , and improved computational speed by 52.64 $$\%$$ compared to the classical KIND.
format Article
id doaj-art-403ed3a3af2d4fc98b1b99273cdde8af
institution OA Journals
issn 2045-2322
language English
publishDate 2024-11-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-403ed3a3af2d4fc98b1b99273cdde8af2025-08-20T02:08:24ZengNature PortfolioScientific Reports2045-23222024-11-0114111610.1038/s41598-024-81706-2Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancementJunfeng Wang0Shenghui Huang1Zhanqiang Huo2Shan Zhao3Yingxu Qiao4School of Software, Henan Polytechnic UniversitySchool of Software, Henan Polytechnic UniversitySchool of Software, Henan Polytechnic UniversitySchool of Software, Henan Polytechnic UniversitySchool of Computer Science and Technology, Henan Polytechnic UniversityAbstract Low-light image enhancement aims to enhance the visibility and contrast of low-light images while eliminating complex degradation issues such as noise, artifacts, and color distortions. Most existing low-light image enhancement methods either focus on quality while neglecting computational efficiency or have limited learning and generalization capabilities. To address these issues, we propose a Bilateral Enhancement Network with signal-to-noise ratio fusion, called BiEnNet, for lightweight and generalizable low-light image enhancement. Specifically, we design a lightweight Bilateral enhancement module with SNR (Signal-to-Noise Ratio) Fusion (BSF), which serves the SNR map of the input low-light image as the interpolation weights to dynamically fuse global brightness features and local detail features extracted from a bilateral network and achieve differentiated enhancement across different regions. To improve the network’s generalization ability, we propose a Luminance Normalization (LNM) module for preprocessing and a Dual-Exposure Processing (DEP) module for post-processing. LNM divides the channels of input features into luminance-related channels and luminance independent channels, and reduces the inconsistency of the degradation distribution of input low-light images by only normalizing the luminance-related channels. DEP learns overexposure and underexposure corrections simultaneously by employing the ReLU activation function, inverting operation, and residual network, which can improve the robustness of enhancement effects under different exposure conditions while reducing network parameters. Experiments on the LOL-V1 dataset shows BiEnNet significantly increased PSNR by 8.6 $$\%$$ and SSIM by 3.6 $$\%$$ compared to FLW-Net, reduced parameters by 98.78 $$\%$$ , and improved computational speed by 52.64 $$\%$$ compared to the classical KIND.https://doi.org/10.1038/s41598-024-81706-2Low-light Image Enhancement, Lightweight, Generalization ability, SNR Fusion, Normalization
spellingShingle Junfeng Wang
Shenghui Huang
Zhanqiang Huo
Shan Zhao
Yingxu Qiao
Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
Scientific Reports
Low-light Image Enhancement, Lightweight, Generalization ability, SNR Fusion, Normalization
title Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
title_full Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
title_fullStr Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
title_full_unstemmed Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
title_short Bilateral enhancement network with signal-to-noise ratio fusion for lightweight generalizable low-light image enhancement
title_sort bilateral enhancement network with signal to noise ratio fusion for lightweight generalizable low light image enhancement
topic Low-light Image Enhancement, Lightweight, Generalization ability, SNR Fusion, Normalization
url https://doi.org/10.1038/s41598-024-81706-2
work_keys_str_mv AT junfengwang bilateralenhancementnetworkwithsignaltonoiseratiofusionforlightweightgeneralizablelowlightimageenhancement
AT shenghuihuang bilateralenhancementnetworkwithsignaltonoiseratiofusionforlightweightgeneralizablelowlightimageenhancement
AT zhanqianghuo bilateralenhancementnetworkwithsignaltonoiseratiofusionforlightweightgeneralizablelowlightimageenhancement
AT shanzhao bilateralenhancementnetworkwithsignaltonoiseratiofusionforlightweightgeneralizablelowlightimageenhancement
AT yingxuqiao bilateralenhancementnetworkwithsignaltonoiseratiofusionforlightweightgeneralizablelowlightimageenhancement