snnTrans-DHZ: A Lightweight Spiking Neural Network Architecture for Underwater Image Dehazing

2025-04-13

AI Summary

Underwater image dehazing suffers from severe visibility degradation due to light scattering and absorption, hindering marine visual operations. To address this, we propose snntTrans-DHZ—a lightweight spiking neural network (SNN)-based dehazing model—marking the first application of SNNs to underwater image restoration. Our method converts static RGB images into temporal spike sequences via time-encoded representation and jointly extracts features in both RGB and LAB color spaces. We design a K-estimator, a joint background-light estimator, and a soft reconstruction module to achieve energy-efficient, high-fidelity dehazing. Training employs surrogate-gradient backpropagation through time (BPTT) with a custom composite loss function. Evaluated on UIEB and EUVP benchmarks, snntTrans-DHZ achieves PSNR of 21.68 dB and 23.46 dB, SSIM of 0.8795 and 0.8439, respectively, with only 0.567M parameters and an energy consumption of 0.0151 J—substantially outperforming state-of-the-art methods.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Learning Theory

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments

Abstract

Underwater image dehazing is critical for vision-based marine operations because light scattering and absorption can severely reduce visibility. This paper introduces snnTrans-DHZ, a lightweight Spiking Neural Network (SNN) specifically designed for underwater dehazing. By leveraging the temporal dynamics of SNNs, snnTrans-DHZ efficiently processes time-dependent raw image sequences while maintaining low power consumption. Static underwater images are first converted into time-dependent sequences by repeatedly inputting the same image over user-defined timesteps. These RGB sequences are then transformed into LAB color space representations and processed concurrently. The architecture features three key modules: (i) a K estimator that extracts features from multiple color space representations; (ii) a Background Light Estimator that jointly infers the background light component from the RGB-LAB images; and (iii) a soft image reconstruction module that produces haze-free, visibility-enhanced outputs. The snnTrans-DHZ model is directly trained using a surrogate gradient-based backpropagation through time (BPTT) strategy alongside a novel combined loss function. Evaluated on the UIEB benchmark, snnTrans-DHZ achieves a PSNR of 21.68 dB and an SSIM of 0.8795, and on the EUVP dataset, it yields a PSNR of 23.46 dB and an SSIM of 0.8439. With only 0.5670 million network parameters, and requiring just 7.42 GSOPs and 0.0151 J of energy, the algorithm significantly outperforms existing state-of-the-art methods in terms of efficiency. These features make snnTrans-DHZ highly suitable for deployment in underwater robotics, marine exploration, and environmental monitoring.

Problem

• Enhancing underwater image visibility degraded by scattering and absorption
• Designing a lightweight SNN for efficient underwater image dehazing
• Reducing power consumption while processing time-dependent image sequences

Innovation

• Lightweight SNN for underwater image dehazing
• RGB to LAB color space transformation
• Surrogate gradient-based BPTT training