基于梯度直方图变换的低照度图像边缘细节增强处理系统  

Low illumination image edge detail enhancement processing system based on gradient histogram transform

作　　者：王铭勋 张晓辉[1] 于心俊[1] 申慧男 WANG Mingxun;ZHANG Xiaohui;YU Xinjun;SHEN Huinan(College of Electrical Engineering,Henan University of Technology,Zhengzhou 450001,China)

机构地区：[1]河南工业大学电气工程学院,河南郑州450001

出　　处：《现代电子技术》2025年第5期75-78,共4页Modern Electronics Technique

基　　金：河南省自然科学基金项目(202300410117);河南省科技攻关项目(192102210240)。

摘　　要：低照度图像的清晰度低且边缘细节信息处理难度大,导致图像视觉质量低。针对这一问题,文中设计一种基于梯度直方图变换的低照度图像边缘细节增强处理系统。首先,利用指数派生函数,校正低照度图像的亮度通道,并调节低照度图像亮度;然后,将高斯函数作为梯度直方图变换的规定化函数,提取已校正亮度通道图像的边缘和纹理信息,进行梯度直方图变换;最后,选取Retinex算法和高斯卷积函数,根据低照度图像的梯度直方图变换结果估计图像的入射光线,实现低照度图像的边缘细节增强处理。系统测试结果表明,该系统能够显著增强低照度图像的边缘细节,提升图像的视觉质量,增强处理后图像的信息熵均高于6。The low sharpness of low illumination images and the difficulties in processing their edge detail information re-sult in low visual quality.In view of the above,this paper designs a low illumination image edge detail enhancement processing system based on gradient histogram transform.Firstly,an exponential derivative function is used to correct the brightness channel of the low illumination images and adjust their brightness.Then,the Gaussian function is used as the regularization function for gradient histogram transform,extracting the edge and texture information of the images with corrected brightness channels,and performing gradient histogram transform.Finally,the Retinex algorithm and Gaussian convolution function are selected to esti-mate the image incident light based on the gradient histogram transform results of the low illumination images,so as to achieve the edge detail enhancement processing of the low illumination images.The system test results show that the system can signifi-cantly enhance the edge details of the low illumination images,improve the visual quality of the images,and the information en-tropy of the enhanced images is higher than 6.

关 键 词：梯度直方图 低照度图像 边缘细节 增强处理 指数派生函数 RETINEX算法 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP399[电子电信—信息与通信工程]