MambaRA-GAN: Underwater Image Enhancement via Mamba and Intra-Domain Reconstruction Autoencoder
Open Access
- 10 September 2025
- journal article
- research article
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 13 (9) , 1745
Abstract
Underwater images frequently suffer from severe quality degradation due to light attenuation and scattering effects, manifesting as color distortion, low contrast, and detail blurring. These issues significantly impair the performance of downstream tasks. Therefore, underwater image enhancement (UIE) becomes a key technology to solve underwater image degradation. However, existing data-driven UIE methods typically rely on difficult-to-acquire paired data for training, severely limiting their practical applicability. To overcome this limitation, this study proposes MambaRA-GAN, a novel unpaired UIE framework built upon a CycleGAN architecture, which introduces a novel integration of Mamba and intra-domain reconstruction autoencoders. The key innovations of our work are twofold: (1) We design a generator architecture based on a Triple-Gated Mamba (TG-Mamba) block. This design dynamically allocates feature channels to three parallel branches via learnable weights, achieving optimal fusion of CNN’s local feature extraction capabilities and Mamba’s global modeling capabilities. (2) We construct an intra-domain reconstruction autoencoder, isomorphic to the generator, to quantitatively assess the quality of reconstructed images within the cycle consistency loss. This introduces more effective structural information constraints during training. The experimental results demonstrate that the proposed method achieves significant improvements across five objective performance metrics. Visually, it effectively restores natural colors, enhances contrast, and preserves rich detail information, robustly validating its efficacy for the UIE task.Keywords
Funding Information
- Yunnan Provincial Major Science and Technology Project (202102AD080002)
This publication has 32 references indexed in Scilit:
- LDNet: High Accuracy Fish Counting Framework using Limited training samples with Density map generation NetworkJournal of King Saud University Computer and Information Sciences, 2024
- Single and multiphase flow leak detection in onshore/offshore pipelines and subsurface sequestration sites: An overviewJournal of Loss Prevention in the Process Industries, 2024
- MuLA-GAN: Multi-Level Attention GAN for Enhanced Underwater VisibilityEcological Informatics, 2024
- Underwater vision enhancement technologies: a comprehensive review, challenges, and recent trendsApplied Intelligence, 2022
- Underwater scene prior inspired deep underwater image and video enhancementPattern Recognition, 2020
- Underwater image enhancement using blending of CLAHE and percentile methodologiesMultimedia Tools and Applications, 2018
- Automatic Red-Channel underwater image restorationJournal of Visual Communication and Image Representation, 2015
- Contrast Limited Adaptive Histogram Equalization1994
- A spatial processor model for object colour perceptionJournal of the Franklin Institute, 1980
- Image enhancement by histogram transformationComputer Graphics and Image Processing, 1977