Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study

Underwater remote sensing image enhancement is complicated by low illumination, color bias, and blurriness, affecting deep-sea monitoring and marine resource development. This study compares a multi-scale fusion-enhanced physical model and deep learning algorithms to optimize intelligent processing....

Full description

Saved in:
Bibliographic Details
Main Authors: Yunsheng Ma, Yanan Cheng, Dapeng Zhang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Journal of Marine Science and Engineering
Subjects:
Online Access:https://www.mdpi.com/2077-1312/13/5/899
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849326961361420288
author Yunsheng Ma
Yanan Cheng
Dapeng Zhang
author_facet Yunsheng Ma
Yanan Cheng
Dapeng Zhang
author_sort Yunsheng Ma
collection DOAJ
description Underwater remote sensing image enhancement is complicated by low illumination, color bias, and blurriness, affecting deep-sea monitoring and marine resource development. This study compares a multi-scale fusion-enhanced physical model and deep learning algorithms to optimize intelligent processing. The physical model, based on the Jaffe–McGlamery model, integrates multi-scale histogram equalization, wavelength compensation, and Laplacian sharpening, using cluster analysis to target enhancements. It performs well in shallow, stable waters (turbidity < 20 NTU, depth < 10 m, PSNR = 12.2) but struggles in complex environments (turbidity > 30 NTU). Deep learning models, including water-net, UWCNN, UWCycleGAN, and U-shape Transformer, excel in dynamic conditions, achieving UIQM = 0.24, though requiring GPU support for real-time use. Evaluated on the UIEB dataset (890 images), the physical model suits specific scenarios, while deep learning adapts better to variable underwater settings. These findings offer a theoretical and technical basis for underwater image enhancement and support sustainable marine resource use.
format Article
id doaj-art-bee26b33bb32498a8ec73810c91b4204
institution Kabale University
issn 2077-1312
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Journal of Marine Science and Engineering
spelling doaj-art-bee26b33bb32498a8ec73810c91b42042025-08-20T03:48:01ZengMDPI AGJournal of Marine Science and Engineering2077-13122025-04-0113589910.3390/jmse13050899Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative StudyYunsheng Ma0Yanan Cheng1Dapeng Zhang2Ship and Maritime College, Guangdong Ocean University, Zhanjiang 524005, ChinaTaizhou Institute of Science & Technology, College of Business, NJUST, Taizhou 225300, ChinaShip and Maritime College, Guangdong Ocean University, Zhanjiang 524005, ChinaUnderwater remote sensing image enhancement is complicated by low illumination, color bias, and blurriness, affecting deep-sea monitoring and marine resource development. This study compares a multi-scale fusion-enhanced physical model and deep learning algorithms to optimize intelligent processing. The physical model, based on the Jaffe–McGlamery model, integrates multi-scale histogram equalization, wavelength compensation, and Laplacian sharpening, using cluster analysis to target enhancements. It performs well in shallow, stable waters (turbidity < 20 NTU, depth < 10 m, PSNR = 12.2) but struggles in complex environments (turbidity > 30 NTU). Deep learning models, including water-net, UWCNN, UWCycleGAN, and U-shape Transformer, excel in dynamic conditions, achieving UIQM = 0.24, though requiring GPU support for real-time use. Evaluated on the UIEB dataset (890 images), the physical model suits specific scenarios, while deep learning adapts better to variable underwater settings. These findings offer a theoretical and technical basis for underwater image enhancement and support sustainable marine resource use.https://www.mdpi.com/2077-1312/13/5/899underwater remote sensingdeep learning algorithmsmulti-scale fusion-enhanced physical modelunderwater image enhancement
spellingShingle Yunsheng Ma
Yanan Cheng
Dapeng Zhang
Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
Journal of Marine Science and Engineering
underwater remote sensing
deep learning algorithms
multi-scale fusion-enhanced physical model
underwater image enhancement
title Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
title_full Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
title_fullStr Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
title_full_unstemmed Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
title_short Comparative Analysis of Traditional and Deep Learning Approaches for Underwater Remote Sensing Image Enhancement: A Quantitative Study
title_sort comparative analysis of traditional and deep learning approaches for underwater remote sensing image enhancement a quantitative study
topic underwater remote sensing
deep learning algorithms
multi-scale fusion-enhanced physical model
underwater image enhancement
url https://www.mdpi.com/2077-1312/13/5/899
work_keys_str_mv AT yunshengma comparativeanalysisoftraditionalanddeeplearningapproachesforunderwaterremotesensingimageenhancementaquantitativestudy
AT yanancheng comparativeanalysisoftraditionalanddeeplearningapproachesforunderwaterremotesensingimageenhancementaquantitativestudy
AT dapengzhang comparativeanalysisoftraditionalanddeeplearningapproachesforunderwaterremotesensingimageenhancementaquantitativestudy