混合型强迫不变的遥感影像薄云、雾去除方法  

Mixed Forcing Invariant Method for Removing Thin Cloud and Fog from Remote Sensing Imagery

作　　者：尹展[1] YIN Zhan(China Non-ferrous Metals Resource Geological Survey, Beijing 100012,China)

机构地区：[1]有色金属矿产地质调查中心,北京100012

出　　处：《遥感信息》2020年第5期31-36,共6页Remote Sensing Information

基　　金：中国地质调查局地质调查项目(DD20190705)。

摘　　要：针对影像判读中云雾干扰去除难的问题,基于大气散射理论,提出一种“HSI颜色空间变换+强迫不变”混合型强迫不变遥感影像薄云、雾去除方法。首先利用Landsat-8卷云波段提取云雾因子;然后进行HSI颜色空间变换,计算出I分量与云雾因子的相关曲线;通过曲线拟合、滤波以及平滑,最终平化为某一值,使得I分量光谱值不随云雾因子值变化,与云雾因子无相关性,即强迫不变;最后依据像元分解理论分离出像素中的薄云、雾光谱贡献量,达到去除云雾目的。实验表明,该方法无须传输参数,处理方便,处理后影像地物轮廓显现,清晰度提高,效果较好。As it is difficult to remove the cloud interference in image interpretation,based on the theory of atmospheric scattering,a hybrid method of“HSI color space transformation+forced invariant”is proposed to remove thin cloud and fog from remote sensing imagery.Firstly,the cloud factor is extracted from Landsat-8 cirrus cloud band.Then,the HSI color space transformation is carried out to calculate the correlation curve between I component and cloud factor.Through curve fitting,filtering and smoothing,the spectral value of I component is finally flattened to a certain value,so that the spectral value of I component does not change with the value of cloud factor,and has no correlation with cloud factor,that is,it is forced to remain unchanged.Finally,the spectral value of I component is separated according to the pixel decomposition theory.The contribution of thin cloud and fog spectrum in pixels can achieve the purpose of removing cloud and fog.The experimental results show that the method does not need to transmit parameters and is easy to process.After processing,the image contour appears,the clarity is improved and the effect is good.

关 键 词：混合型强迫不变 HSI颜色空间变换 影像云雾去除 云雾因子 物理模型 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]