液晶与显示2026，Vol.41Issue(3)：452-461,10.DOI:10.37188/CJLCD.2026-0007
霜冰优化算法与多尺度信息融合的水下图像增强
Rime optimization algorithm and multi-scale information fusion for underwater image enhancement
摘要
Abstract
To address the issues of color distortion and severe loss of details in underwater images,an underwater image enhancement method based on the rime optimization algorithm(RIME)and multi-scale information fusion is proposed.Firstly,to tackle the problem of color distortion,an innovative adaptive gamma correction red channel compensation framework is introduced,which integrates the Dark Channel Prior(DCP)algorithm and is optimized using RIME,ensuring natural color restoration for underwater images.Secondly,the Contrast Limited Adaptive Histogram Equalization(CLAHE)method is applied to enhance image contrast.Finally,the enhanced results from the two channels are processed through a four-weight Gaussian-Laplacian pyramid and fused at multiple scales,ultimately producing images of superior visual quality.Experimental results show that the average values of UIQM,UCIQE,and SSIM achieved by this algorithm reach 4.598,0.633,and 0.726,respectively.Additionally,comparative experiments conducted on the UIEB,RUIE,and EUVP datasets demonstrate that the proposed method significantly outperforms other comparison algorithms,confirming its superior effectiveness in improving the quality of underwater images.关键词
暗通道先验/伽马矫正/霜冰优化算法/多尺度融合Key words
dark channel prior/gamma correction/rime optimization algorithm/multi-scale fusion分类
信息技术与安全科学引用本文复制引用
张爱华,王莹..霜冰优化算法与多尺度信息融合的水下图像增强[J].液晶与显示,2026,41(3):452-461,10.基金项目
辽宁省教育厅科学计划重点支持项目(No.JYTZD2023176)Supported by Education Committee Project of Liaoning(No.JYTZD2023176) （No.JYTZD2023176）
