分割映射的单幅彩色图像去雾方法

针对传统暗通道先验去雾方法在进行大气散射函 数估计时容易出现块状模糊效应的问题,提出了一种 基于分割映射的单幅彩色图像去雾方法。首先对前端采集图像进行近景与远景区域分 割,并基于分 割区域进行亮度信息的分段映射,通过分段计算获取大气散射函数的预测估计值；接着,采 用传统的导向 滤波方法对大气散射函数的估计值进行优化分析,进一步增强图像的边缘信息,改善在大面 积天空颜色情 况下图像边缘的块状模糊效应,提升含雾图像在突变区域的去雾效果。针对实际采集 的含雾图像进 行去雾效果分析和对比,分别基于图像的对比度改善量e、色彩自然度(CN I)、颜色丰富程度(CCI)以及计算耗时等4方面进行定量对比。分析结果表明, 本文方法很好地改善了图像的去雾效果,并进一步提升了运行的实时性。

Single color image dehazing method based on image segmentation

In order to reduce the block fuzzy effect in estimating the atmospheric scattering function based on the traditional color image dehazing method,this paper propos es a single color image dehazing method based on segmentation mapping.First of a ll,the front end image blurring vision and regional segmentation are adopted,and the at mospheric scattering function prediction value is obtained based on the segmentation mapping through the image bright ness information.Second,the orientation of the traditional filtering method is used to estimate the atmospheric scattering function value and to optimize the analysis,which can fur ther enhance the edge information of the image,improve the color of the sky in a large area under the massive image edge blur effect,and enhance the image in the region containing t he fog mutation to fog effect.Finally,the algorithm performance is analyzed through exper iment according to the actual acquisition of the fog image containing the defogging effect analysis and comparison, respectively.The quantitative comparison is conducted based on the image improv ing quantity,the degree of naturalness index,the color colorfulness index and computing time.The results show that this method can improve the effect of fog removing and the re al-time performance.