摘要
雾天拍摄时,图像由于景物受到雾的散射,会因衰减作用而变得模糊不清。为得到相对清晰的图像,提出一种改进的Retinex雾天图像清晰化算法,先将图像进行MSR处理,再进行幂次变换,较大程度地压缩图像动态范围。针对出现的高光区域采用非线性变换来抑制,最后进行维纳滤波去除噪声,增强图像细节,得到清晰化的图像。实验结果表明:该算法能有效改善雾天拍摄图像的清晰度,更适宜人的视觉感受。
When shooting in fog,the scenery is blurred by the fog. In order to get a relatively clear image,an improved Retinex image restoration algorithm was proposed. First,the image was processed by MSR,and then a power transformation was processed,which had a greater degree of compression of image dynamic range. We suppressed high light area appeared through nonlinear transformation and finally we used Wiener filter to remove noise and enhance image detail,and goet clear images.Experiments show that this algorithm can effectively improve the contrast and resolution of fog image.
出处
《重庆理工大学学报（自然科学）》
CAS
2016年第10期118-121,共4页
Journal of Chongqing University of Technology：Natural Science
基金
安徽省教育厅自然科学基金重大项目(KY2014ZD004)
关键词
图像清晰化
MSR
幂次变换
非线性变换
维纳滤波
image restoration
MSR
power transformation
nonlinear transformation
Wiener filter
