Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement

Underwater image enhancement (UIE) serves as a fundamental preprocessing step in ocean remote sensing applications, encompassing marine life detection, archaeological surveying, and subsea resource exploration. However, UIE encounters substantial technical challenges due to the intricate physics of...

Full description

Saved in:
Bibliographic Details
Main Authors: Fei Li, Li Wan, Jiangbin Zheng, Lu Wang, Yue Xi
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/5/759
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850222764241190912
author Fei Li
Li Wan
Jiangbin Zheng
Lu Wang
Yue Xi
author_facet Fei Li
Li Wan
Jiangbin Zheng
Lu Wang
Yue Xi
author_sort Fei Li
collection DOAJ
description Underwater image enhancement (UIE) serves as a fundamental preprocessing step in ocean remote sensing applications, encompassing marine life detection, archaeological surveying, and subsea resource exploration. However, UIE encounters substantial technical challenges due to the intricate physics of underwater light propagation and the inherent homogeneity of aquatic environments. Images captured underwater are significantly degraded through wavelength-dependent absorption and scattering processes, resulting in color distortion, contrast degradation, and illumination irregularities. To address these challenges, we propose a contrastive feature disentanglement network (CFD-Net) that systematically addresses underwater image degradation. Our framework employs a multi-stream decomposition architecture with three specialized decoders to disentangle the latent feature space into components associated with degradation and those representing high-quality features. We incorporate hierarchical contrastive learning mechanisms to establish clear relationships between standard and degraded feature spaces, emphasizing intra-layer similarity and inter-layer exclusivity. Through the synergistic utilization of internal feature consistency and cross-component distinctiveness, our framework achieves robust feature extraction without explicit supervision. Compared to existing methods, our approach achieves a 12% higher UIQM score on the EUVP dataset and outperforms other state-of-the-art techniques on various evaluation metrics such as UCIQE, MUSIQ, and NIQE, both quantitatively and qualitatively.
format Article
id doaj-art-8b18d2fc4d0841549773b62d9bf503d0
institution OA Journals
issn 2072-4292
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-8b18d2fc4d0841549773b62d9bf503d02025-08-20T02:06:13ZengMDPI AGRemote Sensing2072-42922025-02-0117575910.3390/rs17050759Contrastive Feature Disentanglement via Physical Priors for Underwater Image EnhancementFei Li0Li Wan1Jiangbin Zheng2Lu Wang3Yue Xi4School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaHangzhou Institute of Technology, Xidian University, Hangzhou 311200, ChinaSchool of Software and Microelectronics, Northwestern Polytechnical University, Xi’an 710071, ChinaSchool of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaGuangzhou Institute of Technology, Xidian University, Guangzhou 510555, ChinaUnderwater image enhancement (UIE) serves as a fundamental preprocessing step in ocean remote sensing applications, encompassing marine life detection, archaeological surveying, and subsea resource exploration. However, UIE encounters substantial technical challenges due to the intricate physics of underwater light propagation and the inherent homogeneity of aquatic environments. Images captured underwater are significantly degraded through wavelength-dependent absorption and scattering processes, resulting in color distortion, contrast degradation, and illumination irregularities. To address these challenges, we propose a contrastive feature disentanglement network (CFD-Net) that systematically addresses underwater image degradation. Our framework employs a multi-stream decomposition architecture with three specialized decoders to disentangle the latent feature space into components associated with degradation and those representing high-quality features. We incorporate hierarchical contrastive learning mechanisms to establish clear relationships between standard and degraded feature spaces, emphasizing intra-layer similarity and inter-layer exclusivity. Through the synergistic utilization of internal feature consistency and cross-component distinctiveness, our framework achieves robust feature extraction without explicit supervision. Compared to existing methods, our approach achieves a 12% higher UIQM score on the EUVP dataset and outperforms other state-of-the-art techniques on various evaluation metrics such as UCIQE, MUSIQ, and NIQE, both quantitatively and qualitatively.https://www.mdpi.com/2072-4292/17/5/759contrastive feature disentanglementphysical priorsunderwater image enhancement
spellingShingle Fei Li
Li Wan
Jiangbin Zheng
Lu Wang
Yue Xi
Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
Remote Sensing
contrastive feature disentanglement
physical priors
underwater image enhancement
title Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
title_full Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
title_fullStr Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
title_full_unstemmed Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
title_short Contrastive Feature Disentanglement via Physical Priors for Underwater Image Enhancement
title_sort contrastive feature disentanglement via physical priors for underwater image enhancement
topic contrastive feature disentanglement
physical priors
underwater image enhancement
url https://www.mdpi.com/2072-4292/17/5/759
work_keys_str_mv AT feili contrastivefeaturedisentanglementviaphysicalpriorsforunderwaterimageenhancement
AT liwan contrastivefeaturedisentanglementviaphysicalpriorsforunderwaterimageenhancement
AT jiangbinzheng contrastivefeaturedisentanglementviaphysicalpriorsforunderwaterimageenhancement
AT luwang contrastivefeaturedisentanglementviaphysicalpriorsforunderwaterimageenhancement
AT yuexi contrastivefeaturedisentanglementviaphysicalpriorsforunderwaterimageenhancement