基于双重域的图像去雾新算法  

A Novel Image Dehazing Algorithm Based on Dual Domains

作　　者：陈剑鹏[1] 毕笃彦[1] 熊磊[1] 王坤[1] 郭永强[1] 

机构地区：[1]空军工程大学航空航天工程学院,西安710038

出　　处：《电光与控制》2015年第6期17-20,63,共5页Electronics Optics & Control

基　　金：国家自然科学基金(61175029;61203268;61202339)

摘　　要：户外场景的图像经常会由于恶劣的天气而降质退化,从而形成雾霾图像。目前为止,大部分基于单幅图像的去雾算法忽略了噪声的影响,因此本文考虑噪声污染建立新的去雾模型,提出一种三阶段去雾新算法。第一阶段,对降质图像进行预处理,消除噪声对图像的干扰;第二阶段,运用中值暗通道优先(MDCP)算法在空域复原预处理图像,增强其全局对比度;第三阶段,在变换域增强图像的局部对比度,进一步恢复图像的细节信息。与仅仅运用空域和没有考虑噪声的算法相比,所提算法显著改善了图像的质量,增加了图像的清晰度。hazy image. Image of the outdoor scene is usually degraded due to poor weather conditions, thus forming a So far, most single image dehazing algorithms ignore the effect of noise. Therefore, in consideration of the noise pollution, a new dehazing model is established, and a three-stage dehazing algorithm is proposed. In the first stage, the degraded image is preprocessed, to eliminate the interference of noise with image. In the second stage, a Median Dark Channel Prior （MDCP） algorithm is used for restoring the preprocessed image in spatial domain, enhancing the global contrast of the image. In the third stage, the local contrast of the image is enhanced in transform domain, for further restoring the details of image. Compared with those methods using the spatial domain only or without taking into consideration of the noise effect, the proposed algorithm can significantly improve the image quality and definition.

关 键 词：图像处理 降质图像 去雾 降质模型 MDCP 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]