多尺度融合图像去雾方法  

An image dehazing method based on multi-scale fusion

作　　者：邱云明[1] 章生冬 范恩[2] 侯能 QIU Yunming;ZHANG Shengdong;FAN En;HOU Neng(College of Physics and Optoelectronic Engineering,Shenzhen University,Shenzhen 518060,Guangdong Province,P.R.China;Department of Computer Science and Engineering,Shaoxing University,Shaoxing 312000,Guangdong Province,P.R.China;School of Computer Science,Yangtze University,Jingzhou 434023,Hubei Province,P.R.China)

机构地区：[1]深圳大学物理与光电工程学院,广东深圳518060 [2]绍兴文理学院计算科学与工程系,浙江绍兴312000 [3]长江大学计算机科学学院,湖北荆州434023

出　　处：《深圳大学学报（理工版）》2024年第5期594-601,共8页Journal of Shenzhen University(Science and Engineering)

基　　金：国家自然科学基金资助项目(62272311);浙江省基础公益研究计划资助项目(LGG22F010004);甘肃省自然科学基金资助项目(20JR5RA378);绍兴文理学院博士科研启动资助项目(20210026,20205048)。

摘　　要：图像去雾能够使视觉系统适应不同的天气状况.为克服传统暗通道先验方法会在物体边界区域形成光晕效应的问题,提出一种用于估计有雾图像透射率的多尺度融合算法.应用不同大小的最小值半径得到多尺度的透射率估计值,再根据局部区域像素具有类似的透射率值这一现象,对透射率图进行多尺度融合,选择小透射图区域中最亮的像素来计算大气光值,最后使用大气散射模型恢复清晰图像.分别从视觉效果和量化指标两个方面,对比所提方法与传统的基于先验和基于深度学习的去雾方法在进行图像去雾后的效果.结果发现,针对4种典型场景,采用本研究算法去雾后的重构图像能够保留更多的结构、细节和颜色信息,避免了过分增强和边缘部分的雾残留问题,视觉效果均优于对比方法;量化指标峰值信噪比和结构相似性均高于对比方法,分别为15.65和0.78.Dehazing is an important preprocessing technique in computer vision.It can make vision system adapt to different weather conditions.Traditional dark channel prior(DCP)based methods ignore the edge between the objects,which result in halo artifacts.To overcome the problem,this paper proposes a multi-scale fusion method to estimate the transmission map.Firstly,we obtain multi-scale transmission layers in a way of different sizes of minimum filter,and then apply multi-scale fusion technique to obtain the transmission map based on the local smooth assumption.Secondly,we obtain the global air-light by selecting the brightest pixel in the smallest transmission map area.Finally,we obtain the final dehazed result in a way of atmospheric scattering model.In the experiment,we compare the results of proposed method with those of different methods with respect to the visual effects and quantitative metrics to illustrate the effectiveness of the proposed method.The results show that in typical four haze images,by applying our proposed method,the visual effects of dehazed results are better than those of compared methods.In quantitative evaluation,we further compare PSNR and SSIM metrics of our method with that of traditional methods and deep learning methods in D-Hazy dataset.The results show that proposed methods can achieve higher scores than those of prior based and deep learning-based methods.

关 键 词：图像处理 图像去雾 暗通道 多尺度 融合方法 透视率图 图像增强 图像恢复 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]