基于Retinex理论的雾天图像增强算法

Foggy Image Enhancement Algorithm Based on Retinex Theory

针对雾天图像对比度低、细节模糊和色调偏灰等问题,提出了一种基于Retinex理论的雾天图像增强算法。将雾天图像转换到HSI空间,对S分量进行线性拉伸,用改进的SSR算法对I分量进行增强处理,转换回RGB空间,用sigmoid函数进行颜色恢复得到增强图像。实验结果表明,提出的算法能有效地提高雾天图像质量,恢复雾天图像色彩,增强效果优于传统算法。

To deal with the problems of low contrast,blurry details and grayish tone in foggy images,a foggy image enhancement algorithm based on Retinex theory is proposed.The foggy image is converted into HSI space,the S component is linearly stretched and the I component is enhanced with improved SSR algorithm.Then the image is converted back into RGB space.Sigmoid function is used to restore color of image and obtain the enhanced image.The experimental results show that the proposed algorithm can effectively improve the quality of foggy image and restore its color,the enhancement effect of the proposed algorithm is better than that of the traditional algorithm.