LM-CycleGAN: Improving Underwater Image Quality Through Learned Perceptual Image Patch Similarity and Multi-Scale Adaptive Fusion Attention

被引:2
作者
Wu, Jiangyan [1 ,2 ]
Zhang, Guanghui [1 ,2 ]
Fan, Yugang [1 ,2 ]
机构
[1] Kunming Univ Sci & Technol, Fac Informat Engn & Automat, Kunming 650500, Peoples R China
[2] Kunming Univ Sci & Technol, Yunnan Key Lab Intelligent Control & Applicat, Kunming 650500, Peoples R China
关键词
underwater image enhancement; cycle-consistent generative adversarial networks; multi-scale adaptive fusion attention; learned perceptual image patch similarity; ENHANCEMENT;
D O I
10.3390/s24237425
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
The underwater imaging process is often hindered by high noise levels, blurring, and color distortion due to light scattering, absorption, and suspended particles in the water. To address the challenges of image enhancement in complex underwater environments, this paper proposes an underwater image color correction and detail enhancement model based on an improved Cycle-consistent Generative Adversarial Network (CycleGAN), named LPIPS-MAFA CycleGAN (LM-CycleGAN). The model integrates a Multi-scale Adaptive Fusion Attention (MAFA) mechanism into the generator architecture to enhance its ability to perceive image details. At the same time, the Learned Perceptual Image Patch Similarity (LPIPS) is introduced into the loss function to make the training process more focused on the structural information of the image. Experiments conducted on the public datasets UIEB and EUVP demonstrate that LM-CycleGAN achieves significant improvements in Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), Average Gradient (AG), Underwater Color Image Quality Evaluation (UCIQE), and Underwater Image Quality Measure (UIQM). Moreover, the model excels in color correction and fidelity, successfully avoiding issues such as red checkerboard artifacts and blurred edge details commonly observed in reconstructed images generated by traditional CycleGAN approaches.
引用
收藏
页数:16
相关论文
共 36 条
[1]  
Anwar S, 2018, Arxiv, DOI arXiv:1807.03528
[2]   MuLA-GAN: Multi-Level Attention GAN for Enhanced Underwater Visibility [J].
Bakht, Ahsan B. ;
Jia, Zikai ;
Din, Muhayy Ud ;
Akram, Waseem ;
Saoud, Lyes Saad ;
Seneviratne, Lakmal ;
Lin, Defu ;
He, Shaoming ;
Hussain, Irfan .
ECOLOGICAL INFORMATICS, 2024, 81
[3]   Low-Cost, Deep-Sea Imaging and Analysis Tools for Deep-Sea Exploration: A Collaborative Design Study [J].
Bell, Katherine L. C. ;
Chow, Jennifer Szlosek ;
Hope, Alexis ;
Quinzin, Maud C. ;
Cantner, Kat A. ;
Amon, Diva J. ;
Cramp, Jessica E. ;
Rotjan, Randi D. ;
Kamalu, Lehua ;
de Vos, Asha ;
Talma, Sheena ;
Buglass, Salome ;
Wade, Veta ;
Filander, Zoleka ;
Noyes, Kaitlin ;
Lynch, Miriam ;
Knight, Ashley ;
Lourenco, Nuno ;
Girguis, Peter R. ;
de Sousa, Joao Borges ;
Blake, Chris ;
Kennedy, Brian R. C. ;
Noyes, Timothy J. ;
McClain, Craig R. .
FRONTIERS IN MARINE SCIENCE, 2022, 9
[4]   Detect concrete cracks based on OTSU algorithm with differential image [J].
Chen, Bo ;
Zhang, Xuan ;
Wang, Ruitao ;
Li, Zhen ;
Deng, Wei .
JOURNAL OF ENGINEERING-JOE, 2019, 2019 (23) :9088-9091
[5]   Underwater Image Enhancement by Wavelength Compensation and Dehazing [J].
Chiang, John Y. ;
Chen, Ying-Ching .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2012, 21 (04) :1756-1769
[6]   PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN With Dual-Discriminators [J].
Cong, Runmin ;
Yang, Wenyu ;
Zhang, Wei ;
Li, Chongyi ;
Guo, Chun-Le ;
Huang, Qingming ;
Kwong, Sam .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2023, 32 :4472-4485
[7]   Transmission Estimation in Underwater Single Images [J].
Drews-, P., Jr. ;
do Nascimento, E. ;
Moraes, F. ;
Botelho, S. ;
Campos, M. .
2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW), 2013, :825-830
[8]   Underwater image enhancement using blending of CLAHE and percentile methodologies [J].
Garg, Diksha ;
Garg, Naresh Kumar ;
Kumar, Munish .
MULTIMEDIA TOOLS AND APPLICATIONS, 2018, 77 (20) :26545-26561
[9]  
Goodfellow IJ, 2014, ADV NEUR IN, V27, P2672
[10]   Airborne Hyperspectral Imaging for Submerged Archaeological Mapping in Shallow Water Environments [J].
Guyot, Alexandre ;
Lennon, Marc ;
Thomas, Nicolas ;
Gueguen, Simon ;
Petit, Tristan ;
Lorho, Thierry ;
Cassen, Serge ;
Hubert-Moy, Laurence .
REMOTE SENSING, 2019, 11 (19)