%0 Journal Article
%T 基于残差模块的Retinex-Net网络的低照度图像增强算法<br>Low Illuminance Image Enhancement Algorithm of Retinex-Net Network Based on Residual Module
%A 高雪
%A 孙兆永
%A 田益民
%A 宋方方
%J Software Engineering and Applications
%P 157-162
%@ 2325-2278
%D 2023
%I Hans Publishing
%R 10.12677/SEA.2023.121016
%X 为解决低光照环境下图片可见度差以及色彩偏差等问题，提出一种基于Retinex-Net网络改进的低照图增强方法。将残差模块加入到分解网络当中，更好地提取图像信息，然后再进行增强、去噪和融合，最后输出增强后图像。经实验证明，本算法能够使增强后的图像亮度有所提高，凸显出细节信息，色彩不失真，更符合人眼的观察效果，从主观感受与客观评价指标各方面均优于原来的网络结构。<br />
In order to solve the problems of poor visibility and color deviation in low light environment, a low light image enhancement method based on Retinex-Net network was proposed. The residual module is added to the decomposition network to better extract image information, and then enhance de-noising and fusion, and finally output the enhanced image. Experiments show that the enhanced low-illuminance image with enhanced brightness, prominent details and small distortion is real and natural, and the algorithm is superior to the original network structure in terms of subjective feeling and objective evaluation indexes.
%K 图像增强，残差网络，Retinex-Net<br>Image Enhancement
%K Residual Network
%K Retinex-Net
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=61966