A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration

Liu, Jianming; Wang, Congzheng; Feng, Chuncheng; Liu, Lei; Gong, Wanqi; Chen, Zhibo; Liao, Libin; Feng, Chang

doi:10.3390/photonics12100975

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration

by

Jianming Liu

^1,2,3,

Congzheng Wang

^1,2,3,*

,

Chuncheng Feng

^1,2,

Lei Liu

^1,2,

Wanqi Gong

^1,2,

Zhibo Chen

^1,2,

Libin Liao

^1,2,3 and

Chang Feng

^1,2,3,*

¹

National Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, China

²

Institute of Optics and Electronics, Chinese Academy of Sciences, Chengdu 610209, China

³

University of Chinese Academy of Sciences, Beijing 100190, China

^*

Authors to whom correspondence should be addressed.

Photonics 2025, 12(10), 975; https://doi.org/10.3390/photonics12100975

Submission received: 29 May 2025 / Revised: 16 September 2025 / Accepted: 18 September 2025 / Published: 30 September 2025

(This article belongs to the Special Issue Advancements in Optical Metrology and Imaging)

Download Versions Notes

Abstract

This work introduces a two-stage framework, named the Distorted underwater image Restoration Network (DR-Net), to address the complex degradation of underwater images caused by turbulence, water flow fluctuations, and optical properties of water. The first stage employs the Distortion Estimation Network (DE-Net), which leverages a fusion of Transformer and U-Net architectures to estimate distortion information from degraded images and focuses on image distortion recovery. Subsequently, the Image Restoration Generative Adversarial Network (IR-GAN) in the second stage utilizes this estimated distortion information to deblur images and regenerate lost details. Qualitative and quantitative evaluations on both synthetic and real-world image datasets demonstrate that DR-Net outperforms traditional methods and restoration strategies from different perspectives, showcasing its broader applicability and robustness. Our approach enables the restoration of underwater images from a single frame, which facilitates the acquisition of marine resources and enhances seabed exploration capabilities.

Keywords: underwater image restoration; transformer; GAN, deep learning

Share and Cite

MDPI and ACS Style

Liu, J.; Wang, C.; Feng, C.; Liu, L.; Gong, W.; Chen, Z.; Liao, L.; Feng, C. A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration. Photonics 2025, 12, 975. https://doi.org/10.3390/photonics12100975

AMA Style

Liu J, Wang C, Feng C, Liu L, Gong W, Chen Z, Liao L, Feng C. A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration. Photonics. 2025; 12(10):975. https://doi.org/10.3390/photonics12100975

Chicago/Turabian Style

Liu, Jianming, Congzheng Wang, Chuncheng Feng, Lei Liu, Wanqi Gong, Zhibo Chen, Libin Liao, and Chang Feng. 2025. "A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration" Photonics 12, no. 10: 975. https://doi.org/10.3390/photonics12100975

APA Style

Liu, J., Wang, C., Feng, C., Liu, L., Gong, W., Chen, Z., Liao, L., & Feng, C. (2025). A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration. Photonics, 12(10), 975. https://doi.org/10.3390/photonics12100975

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

A Two-Stage Framework for Distortion Information Estimation and Underwater Image Restoration

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI