摘要
有雾图像的复原问题已经成为了机器视觉领域研究的热点。在传统图像复原理论的基础上,提出一种新的基于模糊聚类规则图像去雾方法,用于恢复被雾影响的降质图像。利用模糊聚类的方法对有雾图像进行预处理,从而获取图像复原过程中的透射率参数。为获取有效的景深信息,将透射率图像与提取到的图像饱和度效果图通过多尺度金字塔细化策略进行有效融合,从而获得图像景深信息。最后,通过提取到的透射率和景深信息实现图像的复原。对比实验效果显示,提出的图像去雾算法能够有效去除图像中的雾信息,同时算法的计算速度相对较快。
The image restoration has become an active topic in machine vision. This paper presents a novel dehazing method based on fuzzy clustering rules for restoring the image degraded by fog. The fuzzy clustering method is used to preprocess the image degraded by fog or haze to obtain the transmission parameters in the process of restoration. The transmission map and the saturation image are fused by multi-scale pyramid thinning strategy for the image depth information. Finally, the image is recovered by extracting the transmission and the image depth information. The experimental results show that the method can effectively remove the fog information in the image and the algorithm has a higher computing speed.
出处
《计算机工程与应用》
CSCD
北大核心
2018年第6期26-29,共4页
Computer Engineering and Applications
基金
辽宁省博士科研启动基金(No.201601349)
辽宁省教育厅科学技术青年项目(No.LQ2017004)
关键词
图像去雾
模糊聚类
透射率
机器视觉
image dehazing
fuzzy cluster
transmission
machine vision
