Unsupervised Multi-Expert Learning Model for Underwater Image Enhancement

被引:6
作者
Liu, Hongmin [1 ,2 ]
Zhang, Qi [1 ,2 ]
Hu, Yufan [1 ,2 ]
Zeng, Hui [3 ]
Fan, Bin [1 ,2 ]
机构
[1] Univ Sci & Technol Beijing, Sch Intelligence Sci & Technol, Beijing 100083, Peoples R China
[2] Univ Sci & Technol Beijing, Inst Artificial Intelligence, Beijing 100083, Peoples R China
[3] Univ Sci & Technol Beijing, Sch Automat & Elect Engn, Beijing 100083, Peoples R China
关键词
Image color analysis; Image enhancement; Imaging; Image edge detection; Degradation; Training; Task analysis; Multi-expert learning; underwater image enhancement; unsupervised learning; WATER;
D O I
10.1109/JAS.2023.123771
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Underwater image enhancement aims to restore a clean appearance and thus improves the quality of underwater degraded images. Current methods feed the whole image directly into the model for enhancement. However, they ignored that the R, G and B channels of underwater degraded images present varied degrees of degradation, due to the selective absorption for the light. To address this issue, we propose an unsupervised multi-expert learning model by considering the enhancement of each color channel. Specifically, an unsupervised architecture based on generative adversarial network is employed to alleviate the need for paired underwater images. Based on this, we design a generator, including a multi-expert encoder, a feature fusion module and a feature fusion-guided decoder, to generate the clear underwater image. Accordingly, a multi-expert discriminator is proposed to verify the authenticity of the R, G and B channels, respectively. In addition, content perceptual loss and edge loss are introduced into the loss function to further improve the content and details of the enhanced images. Extensive experiments on public datasets demon-strate that our method achieves more pleasing results in vision quality. Various metrics (PSNR, SSIM, UIQM and UCIQE) evaluated on our enhanced images have been improved obviously.
引用
收藏
页码:708 / 722
页数:15
相关论文
共 50 条
[41]   Neuromorphic Computing Network for Underwater Image Enhancement and Beyond [J].
Xiao, Fengqi ;
Liu, Jiahui ;
Huang, Yifan ;
Cheng, En ;
Yuan, Fei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62
[42]   Underwater Image Enhancement Based on Multichannel Adaptive Compensation [J].
Qiang, Hu ;
Zhong, Yuzhong ;
Zhu, Yuqi ;
Zhong, Xuke ;
Xiao, Quan ;
Dian, Songyi .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2024, 73 :1-10
[43]   Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding [J].
Li, Chongyi ;
Anwar, Saeed ;
Hou, Junhui ;
Cong, Runmin ;
Guo, Chunle ;
Ren, Wenqi .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2021, 30 :4985-5000
[44]   Feature distillation and guide network for unsupervised underwater image enhancement [J].
Luan, Xin ;
Wang, Qiang ;
Fan, Huijie ;
Chen, Xi'ai ;
Han, Zhi ;
Tang, Yandong .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 136
[45]   Semi-DinatNet: Two-Stage Underwater Image Enhancement With Semi-Supervised Learning [J].
Ye, Renchuan ;
Huang, Xinming ;
Qian, Yuqiang ;
Zhang, Zhihao .
IEEE ACCESS, 2024, 12 :151236-151250
[46]   Unsupervised Learning for 2D Image Texture Enhancement [J].
Labinghisa, Boney ;
Kim, Jeong-Su ;
Lee, Dong Myung .
12TH INTERNATIONAL CONFERENCE ON ICT CONVERGENCE (ICTC 2021): BEYOND THE PANDEMIC ERA WITH ICT CONVERGENCE INNOVATION, 2021, :587-589
[47]   Lighting the darkness in the sea: A deep learning model for underwater image enhancement [J].
Xie, Yaofeng ;
Yu, Zhibin ;
Yu, Xiao ;
Zheng, Bing .
FRONTIERS IN MARINE SCIENCE, 2022, 9
[48]   A Hybrid Framework for Underwater Image Enhancement [J].
Li, Xinjie ;
Hou, Guojia ;
Tan, Lu ;
Liu, Wanquan .
IEEE ACCESS, 2020, 8 (08) :197448-197462
[49]   FDCE-Net: Underwater Image Enhancement With Embedding Frequency and Dual Color Encoder [J].
Cheng, Zheng ;
Fan, Guodong ;
Zhou, Jingchun ;
Gan, Min ;
Chen, C. L. Philip .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2025, 35 (02) :1728-1744
[50]   UWMamba: UnderWater Image Enhancement With State Space Model [J].
An, Guanhua ;
He, Ao ;
Wang, Yudong ;
Guo, Jichang .
IEEE SIGNAL PROCESSING LETTERS, 2024, 31 :2725-2729