UIESNN: A Scale-Aware Spiking Network for Underwater Image Enhancement

2026-05-08

AI Summary

This work addresses the large-scale low-frequency degradation in underwater images caused by wavelength-dependent color casts and scattering, a challenge that existing spiking neural networks (SNNs) struggle to resolve due to their inherently local receptive fields, which hinder globally consistent enhancement. To overcome this limitation, we propose UIESNN, a scale-aware fully spiking network for underwater image enhancement, which innovatively integrates multi-scale pooling responses into the membrane potential dynamics of Leaky Integrate-and-Fire (LIF) neurons to establish a heterogeneous scale-dependent activation mechanism. Furthermore, our approach pioneers end-to-end frequency decomposition and attention-based refinement within an SNN framework, combined with a spiking residual architecture to simultaneously enhance global consistency and preserve fine details. Evaluated on the EUVP and LSUI benchmarks, UIESNN achieves state-of-the-art performance among SNN-based methods, significantly improving color fidelity and spatial coherence while maintaining low energy consumption.

Abstract

Underwater image enhancement (UIE) is a practically important yet underexplored application of spiking neural networks (SNNs), where the dominant degradations are large-scale and low-frequency, such as wavelength-dependent colour casts and scattering-induced veiling. Existing SNN restoration designs rely on locally bounded spiking perception, which can limit global correction and lead to saturated or inconsistent representations. To address these challenges, we propose a scale-aware SNN framework for UIE named UIESNN. At its core is a Multi-scale Pooling LIF Block (MPLB) that injects hierarchical multi-scale pooling responses into membrane dynamics, thereby enlarging the effective receptive field while preserving fine-grained details and inducing heterogeneous scale-dependent activations. Building on MPLB, we design a spiking residual architecture that integrates frequency decomposition and attention-based refinement in a fully spike-driven pipeline. Extensive experiments on the EUVP and LSUI benchmarks demonstrate that UIESNN achieves state-of-the-art performance among SNN-based methods, delivering improved colour fidelity and spatial coherence with competitive energy cost.

Problem

• underwater image enhancement
• spiking neural networks
• scale-aware
• colour casts
• veiling

Innovation

• scale-aware
• spiking neural networks
• multi-scale pooling
• underwater image enhancement
• membrane dynamics