基于Retinex的可变注意力低照度水下图像增强  

Variable attention low illumination underwater image enhancement based on Retinex

作　　者：陶洋[1] 龚霁霆 周立群 TAO Yang;GONG Jiting;ZHOU Liqun(College of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学通信与信息工程学院,重庆400065

出　　处：《液晶与显示》2025年第3期481-492,共12页Chinese Journal of Liquid Crystals and Displays

基　　金：国家重点研发计划(No.2019YFB2102001)。

摘　　要：针对低照度水下图像增强任务存在光照不足和水下散射引起的复杂退化问题,提出了一种基于Retinex的可变注意力低照度水下图像增强框架。设计了光照引导的可变注意力模块,利用亮度特征图提供的语义信息帮助增强暗区域,提高模型的自适应增强能力。构建了空频域特征融合模块,通过多尺度特征提取和不同层次间的特征融合,提升图像纹理信息的显著性。引入介质传输模块,进一步解决增强过程中水下散射引起的图像伪影。实验结果表明,本文方法相较其他前沿方法在4种客观指标上均有显著提升,PSNR达到22.7818,SSIM达到0.8821,UCIQE达到0.6146,UIQM达到3.3659。增强后的图像视觉质量出色,验证了该算法提升水下低照度图像清晰度的有效性。To address issues of insufficient illumination and complex degradation caused by underwater scattering in low-light underwater image enhancement tasks,we propose a Retinex-based variable attention framework.An illumination-guided variable attention module is designed,which uses semantic information from brightness feature maps to enhance dark regions and improve the adaptive enhancement capability of the model.We construct a spatial-frequency domain feature fusion module,which enhances the prominence of image texture information through multi-scale feature extraction and inter-level feature fusion.A medium transmission module is introduced to further address image artifacts caused by underwater scattering during enhancement.The experimental results show that compared with other cutting-edge methods,our method has significant improvements in four objective indicators,with PSNR reaching 22.7818,SSIM reaching 0.8821,UCIQE reaching 0.6146,and UIQM reaching 3.3659.The enhanced images exhibit excellent visual quality,validating the effectiveness of our algorithm in improving the clarity of low-light underwater images.

关 键 词：水下图像增强 低照度 RETINEX 可变注意力 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]