Super resolution of underwater image based on generating generative adversarial networks

Yunting Lai; Zhuang Zhou; Binghua Su; Jiongjiang Chen; Wanxin Liang; Chenhao Ma; Tenghui Wang; Zexin Zheng

doi:10.1117/12.2652062

20 January 2023 Super resolution of underwater image based on generating generative adversarial networks

Yunting Lai, Zhuang Zhou, Binghua Su, Jiongjiang Chen, Wanxin Liang, Chenhao Ma, Tenghui Wang, Zexin Zheng

Proceedings Volume 12561, AOPC 2022: Atmospheric and Environmental Optics; 125610E (2023) https://doi.org/10.1117/12.2652062
Event: Applied Optics and Photonics China 2022 (AOPC2022), 2022, Beijing, China

Abstract

High-quality underwater images and videos play an important role for exploitation tasks in underwater environments, but the complexity of the underwater imaging environment makes the quality of the acquired underwater images generally low. In order to improve underwater image quality by enhancing the resolution of underwater images, we propose an underwater image super-resolution method based on the improvement of SRGAN. The images generated by conventional super-resolution methods lose details and high-frequency information, resulting in overly smooth images. We have improved SRGAN by replacing the original Residual Block with Residual Dense Block in the feature extraction network, which improves the network performance and speeds up the training and testing of the model. And the Shuffle Attention mechanism is incorporated after each residual block, which efficiently combines the channel attention mechanism and the spatial attention mechanism, significantly improving the network's ability to extract features and generating high-resolution images with richer detail information. At the same time, our method effectively solves the problem that the generator generates images that are too smooth and without grain details. Our method is compared with SRGAN and SRResnet methods in USR-248, UFO public underwater image dataset, and the experimental results demonstrate that our method generates super-resolution images with better image detail enhancement and higher PSNR and SSIM values.

Citation Download Citation

Yunting Lai, Zhuang Zhou, Binghua Su, Jiongjiang Chen, Wanxin Liang, Chenhao Ma, Tenghui Wang, and Zexin Zheng "Super resolution of underwater image based on generating generative adversarial networks", Proc. SPIE 12561, AOPC 2022: Atmospheric and Environmental Optics, 125610E (20 January 2023); https://doi.org/10.1117/12.2652062

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Super resolution

Data modeling

Image enhancement

Eye models

Image resolution

Network architectures

Performance modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years