一种基于改进导向滤波器的HDR图像分层方法  

A HDR Image Layering Method Based on Improved Guided Filter

作　　者：聂雯萱 严利民[1,2] 

机构地区：[1]上海大学机电工程与自动化学院,上海200444 [2]上海大学微电子研究与开发中心,上海200444

出　　处：《工业控制计算机》2022年第7期87-89,共3页Industrial Control Computer

摘　　要：由于现有的大部分显示设备的动态范围与HDR图像不匹配,无法对HDR图像进行直接输出,因此,对HDR图像映射技术的研究就变得十分重要。在HDR图像映射处理的分层算法中,存在因亮度改变而引起图像边缘区域出现光晕现象以及信息缺失问题,选择导向滤波器对图像进行滤波处理,并针对传统导向滤波器边缘粗化的问题进行了改进,在对基本层图像进行亮度压缩的同时,对细节层图像进行多尺度滤波处理,并根据图像的亮度压缩比例对细节层图像进行适应性调整,有效地改善了输出图像的质量。Since the dynamic range of most of the existing display devices does not match the HDR image,the HDR image cannot be directly output.Therefore,the research on HDR image mapping technology becomes very important.In the layered algorithm of HDR image mapping processing,there is a problem of halo phenomenon and information loss in the edge area of the image caused by the change of brightness.In this paper,the guided filter is selected to filter the image,and the problem of edge coarsening of the traditional guided filter has been improved.While compressing the brightness of the base layer image,multi-scale filtering is performed on the detail layer image,and the detail layer image is adaptively adjusted according to the brightness compression ratio of the image,which effectively improves the quality of the output image.

关 键 词：高动态范围 图像分层 导向滤波 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]