高级检索
杨万挺, 汪荣贵, 方帅, 张璇. 滤波器可变的Retinex雾天图像增强算法[J]. 计算机辅助设计与图形学学报, 2010, 22(6): 965-971.
引用本文: 杨万挺, 汪荣贵, 方帅, 张璇. 滤波器可变的Retinex雾天图像增强算法[J]. 计算机辅助设计与图形学学报, 2010, 22(6): 965-971.
Yang Wanting, Wang Ronggui, Fang Shuai, Zhang Xuan. Variable Filter Retinex Algorithm for Foggy Image Enhancement[J]. Journal of Computer-Aided Design & Computer Graphics, 2010, 22(6): 965-971.
Citation: Yang Wanting, Wang Ronggui, Fang Shuai, Zhang Xuan. Variable Filter Retinex Algorithm for Foggy Image Enhancement[J]. Journal of Computer-Aided Design & Computer Graphics, 2010, 22(6): 965-971.

滤波器可变的Retinex雾天图像增强算法

Variable Filter Retinex Algorithm for Foggy Image Enhancement

  • 摘要: 针对现有中心环绕Retinex图像增强算法滤波器固定,不能对雾天图像有效地同时增强细节与色彩保真的问题,提出一种滤波器可变的Retinex图像增强算法.首先根据雾天图像雾化程度的分布特征得到滤波参数的阈值;然后将原图中每一子块的局部信息与阈值信息作差运算得到相应的滤波器,由此计算出该子块的入射分量,并通过部分重叠策略平移子块完成对整幅图像的入射分量估计;最后将原图减去整幅图像的入射分量得到反射分量,实现对图像的增强.通过主观观测和客观评价的结果表明:该算法比现有的中心环绕Retinex算法在细节增强和颜色保真方面具有更好的效果.

     

    Abstract: A new Retinex algorithm based on variable filters is proposed to improve the existing center-surround Retinex algorithm.The proposed algorithm can enhance details effectively while maintaining the color fidelity in the applications of foggy images enhancement.The existing center-surround Retinex algorithm cannot achieve such function because of the utilizing of fixed filter.First the thresholds are worked out according to the image quality distributing.Then in each sub-block,the algorithm uses the local information and the thresholds to make a corresponding filter to estimate illumination of the current sub-block,and adopt the approach of partially overlapped sub-block to get the whole image's illumination.At last the whole illumination is subtracted from original image.The results show that the proposed algorithm is better than the existing center surround Retinex algorithm in enhancing details and preserving color fidelity.

     

/

返回文章
返回