雾霾天气下多尺度劣质图像质量增强方法研究  

Study on the Quality Enhancement Method of Low-quality Image Quality in Smog Weather

作　　者：焦阳[1] 李鑫[1] 

机构地区：[1]郑州升达经贸管理学院信息工程系,郑州451191

出　　处：《科技通报》2017年第11期133-136,共4页Bulletin of Science and Technology

摘　　要：提出一种基于Retinex算法的多尺度劣质图像增强方法。该方法首先利用图像保持的映射算法进行多尺度劣质图像的特征提取,结合特征提取的图像利用Retinex增强算法对多尺度劣质图像进行图像增强,引入变换实现多尺度劣质图像动态范围的压缩,采用非线性的变换对多尺度劣质图像的高光区域进行抑制,利用反锐化滤波代替高斯函数的滤波和原始的多尺度劣质图像进行卷积,以图像增强的结果完成对雾霾天气下多尺度劣质图像质量增强方法研究。实验结果表明,该方法能有效地提高雾天图像的对比度与饱和度,并保持图像色度不变。In this paper, a new method of image enhancement based on the Retinex algorithm is proposed.This method firstly using image maintain the mapping algorithm of multi-scale inferior image feature extraction, combined with the feature extraction of image using the Retinex of multi-scale inferior image enhancement algorithm for image enhancement, the introduction of transformation to realize the compression of multi-scale inferior image dynamic range, using the nonlinear transformation of multiscale inferior image highlights areas to suppress, use the sharpen filter instead of gaussian function filter and multi-scale inferior image convolution of the original, with the result of image enhancement done for fog weather multi-scale inferior image quality enhancement method research. The experimental results show that the method can improve the contrast degree and saturation of the image and keep the color of the image unchanged.

关 键 词：雾霾天气 多尺度劣质 图像增强 Retinex增强算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]