计算机应用研究2017，Vol.34Issue(4)：1277-1280,4.DOI:10.3969/j.issn.1001-3695.2017.04.073
基于暗通道先验和核回归的图像去雾研究
Research of image dehazing based on darkness-channel prior knowledge and kernel regression
摘要
Abstract
In view of the problem that atmospheric scattering degraded image quality in the hazy weather,this paper proposed a research of image dehazing based on darkness-channel prior knowledge and kernel regression theory.Firstly,it estimated atmospheric light intensity and initial transmission rate in terms of darkness channel prior theory.Secondly,it adopted the kernel regression method to refine initial transmission rate.Finally,it used transmission rate refined and atmospheric light intensity estimated to restore the haze image.According to a large number of experimental data,it shows that this research can remove the fog of image effectively.Compared with the most advanced methods,the image processed by this algorithm not only retains more details of the original image,but also improves the image definition largely.关键词
暗通道先验/核回归/透射率/大气光强度Key words
darkness-channel prior knowledge/kernel regression theory/transmission rate/atmospheric light intensity分类
信息技术与安全科学引用本文复制引用
乔伟伟,谢从华,刘永俊,王晓楠,姚宇峰..基于暗通道先验和核回归的图像去雾研究[J].计算机应用研究,2017,34(4):1277-1280,4.基金项目
国家自然科学基金资助项目(61402204) （61402204）
江苏省自然科学基金资助项目(BK2012209,BK20130529) （BK2012209,BK20130529）
苏州市科技发展计划项目(SYG201409) （SYG201409）
