DEEP-SEA: Deep-Learning Enhancement for Environmental Perception in Submerged Aquatics

2025-08-18

AI Summary

Underwater images suffer from severe degradation—including low contrast, color distortion, and structural blurring—due to light scattering, absorption, and turbidity, significantly hindering marine biodiversity monitoring and environmental perception for autonomous underwater vehicles. To address this, we propose the Dual-Frequency Enhanced Self-Attention Spatial-Frequency Modulator (DFE-SFMod), the first framework to jointly model spatial structure preservation and adaptive frequency-domain feature optimization. DFE-SFMod employs a dual-branch frequency-domain encoder to separately extract low-frequency global semantics and high-frequency details, and introduces a self-attention-driven spatial-frequency interaction mechanism to enforce cross-domain feature consistency. Evaluated on EUVP and LSUI benchmarks, our method surpasses state-of-the-art approaches, achieving average improvements of 1.82 dB in PSNR and 0.032 in SSIM. It effectively restores fine textures and geometric structures, enabling high-fidelity ecological analysis and robust autonomous navigation.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Continuous and reliable underwater monitoring is essential for assessing marine biodiversity, detecting ecological changes and supporting autonomous exploration in aquatic environments. Underwater monitoring platforms rely on mainly visual data for marine biodiversity analysis, ecological assessment and autonomous exploration. However, underwater environments present significant challenges due to light scattering, absorption and turbidity, which degrade image clarity and distort colour information, which makes accurate observation difficult. To address these challenges, we propose DEEP-SEA, a novel deep learning-based underwater image restoration model to enhance both low- and high-frequency information while preserving spatial structures. The proposed Dual-Frequency Enhanced Self-Attention Spatial and Frequency Modulator aims to adaptively refine feature representations in frequency domains and simultaneously spatial information for better structural preservation. Our comprehensive experiments on EUVP and LSUI datasets demonstrate the superiority over the state of the art in restoring fine-grained image detail and structural consistency. By effectively mitigating underwater visual degradation, DEEP-SEA has the potential to improve the reliability of underwater monitoring platforms for more accurate ecological observation, species identification and autonomous navigation.

Problem

• Enhancing underwater image clarity degraded by light scattering and absorption
• Restoring color and structural details in turbid aquatic environments
• Improving marine biodiversity monitoring and autonomous exploration accuracy

Innovation

• Deep learning-based underwater image restoration model
• Dual-Frequency Enhanced Self-Attention Modulator
• Enhances low- and high-frequency image information