基于小波变换的改进暗通道去雾算法  

Image Defogging Algorithm Based on Wavelet Transform and Improved Dark Channel

作　　者：王文科 胡红萍[1] 曹胜芳 WANG Wen-ke;HU Hong-ping;CAO Sheng-fang(School of Mathematics,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学数学学院,太原030051

出　　处：《科学技术与工程》2023年第15期6528-6535,共8页Science Technology and Engineering

基　　金：山西省基础研究计划(20210302123019);山西省回国留学人员科研项目(2020-104,2021-108)。

摘　　要：针对暗通道先验算法在景深较大处会出现颜色失真,且易受噪声干扰和运行时间久等问题,提出了一种基于多尺度小波变换的改进融合暗通道去雾方法。首先对有雾图像作二级小波分解,再对得到的高频分量利用软阈值去噪,对低频分量利用改进的自适应融合暗通道进行去雾。最后利用一个局部线性模型将高低频分量系数关联进行小波重构。实验结果表明,提出的算法具有较高的去雾效率,且能很好地提高去雾图像的质量。Aiming at the problems of color distortion,noise interference and long running time in the dark channel prior algorithm with large depth of field,an image defogging method based on multi-scale wavelet transform and improved fusion dark channel was proposed.Firstly,the foggy image was decomposed by two-level wavelet transform,the high-frequency component was denoised by soft threshold,and the low-frequency component was defogged by improved adaptive fusion dark channel.Finally,a local linear model was used to correlate the high-frequency and low-frequency component coefficients for wavelet reconstruction.The experiment show that the proposed algorithm has high defogging effect and can improve the quality of defogging image.

关 键 词：暗通道先验 小波变换 形态学梯度 图像去雾 透射率 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]