基于国产HISEA-1 SAR与光学影像联合去云研究  

Research on Cloud Removal Based on Combination ofDomestic HISEA-1 SAR and Optical Image

作　　者：李金超 王迎春 马卫春[1,2] 刘慧珺 LI Jinchao;WANG Yingchun;MA Weichun;LIU Huijun(Provincal Fundamental Geomatic Center of Anhui,Hefei 230031,China;Key Laboratory ofJianghuai Arable Land Resources Protection and Eco-restoration,Ministry of Natural Resources of the People’s Republic of China,Hefei 230031,China)

机构地区：[1]安徽省基础测绘信息中心,合肥230031 [2]自然资源部江淮耕地资源保护与生态修复重点实验室,合肥230031

出　　处：《遥感信息》2024年第1期66-72,共7页Remote Sensing Information

基　　金：安徽省测绘局2023年度科技创新项目(2023-KJ-04)。

摘　　要：长期以来,光学影像中的云问题一直是阻碍遥感科学研究与应用的难题与挑战,尤其针对极端气候引发地表结构严重改变的地质灾害应急救援工作,开展遥感影像去云工作有着十分重要的意义。针对遥感影像去云问题,立足于光学影像中云区的物理特征,提出通过主成分分析和频率直方图匹配联合方法对国产HISEA-1 SAR和GF-6多光谱影像联合去云研究,并对去云结果进行了定性和定量的评价。结果表明,这种非同源影像的联合去云方法可行,达到了较为理想的去云效果,不仅有效恢复了云覆盖区域的地物信息,同时增强了地物的纹理结构特征,有效发挥了多源数据间隐含的优势,极大提升了影像的可判读能力,为卫星影像去云工作提供一种新的思路,也为应急救援工作提供一种强有力的保障手段。For a long time,the cloud problem in optical image has been a difficult problem and challenge hindering the research and application of remote sensing science.It is of great significance to carry out cloud removal by remote sensing images,especially for emergency rescue work of geological disasters caused by severe changes in surface structure caused by extreme climate.Aiming at the problem of cloud removal from remote sensing images,based on the physical characteristics of cloud areas in optical images,this paper conducts a joint cloud removal study on domestic HISEA-1 SAR and GF-6 multispectral images through the combined method of principal component analysis and frequency histogram matching.Qualitative and quantitative evaluations are conducted on the results of cloud removal.The results show that this non-homologous joint cloud removal method is feasible and achieves ideal cloud removal effect.This method not only effectively restores the information of the ground objects in the cloud coverage area,but also enhances the texture structural features of the ground objects.It has effectively played the hidden advantages of multi-source data and greatly improved the interpretability of images.It can provide a new idea for satellite image cloud removal,and also provide a powerful guarantee for emergency rescue work.

关 键 词：海丝一号 主成分分析 高分六号 去云 应急保障 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]