Underwater image enhancement via multiscale disentanglement strategy

Abstract Underwater images suffer from color casts, low illumination, and blurred details caused by light absorption and scattering in water. Existing data-driven methods often overlook the scene characteristics of underwater imaging, limiting their expressive power. To address the above issues, we...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiaquan Yan, Hao Hu, Yijian Wang, Muhammad Wasim Nawaz, Naveed Ur Rehman Junejo, Ente Guo, Huibin Feng
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-89109-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850190781123395584
author Jiaquan Yan
Hao Hu
Yijian Wang
Muhammad Wasim Nawaz
Naveed Ur Rehman Junejo
Ente Guo
Huibin Feng
author_facet Jiaquan Yan
Hao Hu
Yijian Wang
Muhammad Wasim Nawaz
Naveed Ur Rehman Junejo
Ente Guo
Huibin Feng
author_sort Jiaquan Yan
collection DOAJ
description Abstract Underwater images suffer from color casts, low illumination, and blurred details caused by light absorption and scattering in water. Existing data-driven methods often overlook the scene characteristics of underwater imaging, limiting their expressive power. To address the above issues, we propose a Multiscale Disentanglement Network (MD-Net) for Underwater Image Enhancement (UIE), which mainly consists of scene radiance disentanglement (SRD) and transmission map disentanglement (TMD) modules. Specifically, MD-Net first disentangles original images into three physical parameters which are scene radiance (clear image), transmission map, and global background light. The proposed network then reconstructs these physical parameters into underwater images. Furthermore, MD-Net introduces class adversarial learning between the original and reconstructed images to supervise the disentanglement accuracy of the network. Moreover, we design a multi-level fusion module (MFM) and dual-layer weight estimation unit (DWEU) for color cast adjustment and visibility enhancement. Finally, we conduct extensive qualitative and quantitative experiments on three benchmark datasets, which demonstrate that our approach outperforms other traditional and state-of-the-art methods. Our code and results are available at: https://github.com/WYJGR/MD-Net .
format Article
id doaj-art-047c174e5486403f8a6a9eeb132a85c8
institution OA Journals
issn 2045-2322
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-047c174e5486403f8a6a9eeb132a85c82025-08-20T02:15:11ZengNature PortfolioScientific Reports2045-23222025-02-0115111510.1038/s41598-025-89109-7Underwater image enhancement via multiscale disentanglement strategyJiaquan Yan0Hao Hu1Yijian Wang2Muhammad Wasim Nawaz3Naveed Ur Rehman Junejo4Ente Guo5Huibin Feng6The Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, School of Computer and Big Data, Minjiang UniversityThe Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, School of Computer and Big Data, Minjiang UniversitySchool of Ophthalmology and Optometry, Wenzhou Medical UniversityDepartment of Computer Engineering, The University of LahoreDepartment of Computer Engineering, The University of LahoreThe Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, School of Computer and Big Data, Minjiang UniversityThe Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, School of Computer and Big Data, Minjiang UniversityAbstract Underwater images suffer from color casts, low illumination, and blurred details caused by light absorption and scattering in water. Existing data-driven methods often overlook the scene characteristics of underwater imaging, limiting their expressive power. To address the above issues, we propose a Multiscale Disentanglement Network (MD-Net) for Underwater Image Enhancement (UIE), which mainly consists of scene radiance disentanglement (SRD) and transmission map disentanglement (TMD) modules. Specifically, MD-Net first disentangles original images into three physical parameters which are scene radiance (clear image), transmission map, and global background light. The proposed network then reconstructs these physical parameters into underwater images. Furthermore, MD-Net introduces class adversarial learning between the original and reconstructed images to supervise the disentanglement accuracy of the network. Moreover, we design a multi-level fusion module (MFM) and dual-layer weight estimation unit (DWEU) for color cast adjustment and visibility enhancement. Finally, we conduct extensive qualitative and quantitative experiments on three benchmark datasets, which demonstrate that our approach outperforms other traditional and state-of-the-art methods. Our code and results are available at: https://github.com/WYJGR/MD-Net .https://doi.org/10.1038/s41598-025-89109-7Underwater image enhancementDisentanglement strategyMultiscale feature fusionUnderwater optical imaging
spellingShingle Jiaquan Yan
Hao Hu
Yijian Wang
Muhammad Wasim Nawaz
Naveed Ur Rehman Junejo
Ente Guo
Huibin Feng
Underwater image enhancement via multiscale disentanglement strategy
Scientific Reports
Underwater image enhancement
Disentanglement strategy
Multiscale feature fusion
Underwater optical imaging
title Underwater image enhancement via multiscale disentanglement strategy
title_full Underwater image enhancement via multiscale disentanglement strategy
title_fullStr Underwater image enhancement via multiscale disentanglement strategy
title_full_unstemmed Underwater image enhancement via multiscale disentanglement strategy
title_short Underwater image enhancement via multiscale disentanglement strategy
title_sort underwater image enhancement via multiscale disentanglement strategy
topic Underwater image enhancement
Disentanglement strategy
Multiscale feature fusion
Underwater optical imaging
url https://doi.org/10.1038/s41598-025-89109-7
work_keys_str_mv AT jiaquanyan underwaterimageenhancementviamultiscaledisentanglementstrategy
AT haohu underwaterimageenhancementviamultiscaledisentanglementstrategy
AT yijianwang underwaterimageenhancementviamultiscaledisentanglementstrategy
AT muhammadwasimnawaz underwaterimageenhancementviamultiscaledisentanglementstrategy
AT naveedurrehmanjunejo underwaterimageenhancementviamultiscaledisentanglementstrategy
AT enteguo underwaterimageenhancementviamultiscaledisentanglementstrategy
AT huibinfeng underwaterimageenhancementviamultiscaledisentanglementstrategy