高分影像地形辐射校正典型模型实验与分析  

Experiments and Analyses of Typical Terrain Radiometric Correction Models on High Resolution Images

作　　者：毛亚萍 吴冲[1] 朱桂海 MAO Yaping WU Chong ZHU Guihai(Remote Sensing and Mapping Institute of Sichuan ,Chengdu 610100,China)

机构地区：[1]四川省遥感信息测绘院,成都610100

出　　处：《遥感信息》2017年第4期126-131,共6页Remote Sensing Information

基　　金：四川省遥感信息测绘院雷达测图与监测技术联合实验室开放基金(SYS2015KFJJ03)

摘　　要：鉴于目前地形辐射校正研究多集中于中低分辨率影像,对高空间分辨率影像的研究较少,对比实验和分析了4种常用地形辐射校正模型的校正效果。以四川省3个不同山区作为实验区,选取WorldView-2、高分一号、GeoEye-1高空间分辨率影像作为源数据,对比实验和分析了C校正模型、矩匹配方法、Shepherd模型和Sandmeier模型等4种常用地形辐射校正模型的校正效果,并通过目视判读和定量分析对校正结果进行评价。结果表明,在缺乏高精度DEM和大气参数的情况下,C校正模型和矩匹配方法2种经验模型的校正结果整体优于Shepherd模型和Sandmeier模型2种物理模型。This paper provides a comprehensive comparison of four classical topographic radiometric correction models. Three experiments for three different mountain areas in Sichuan province were conducted. WorldView-2, GF-1 and GeoEye-1 images were used as the data sources in all the experiments of C correction model, method of moment, Shepherd model and Sandmeier model. Both visual and quantitative assessments were performed. The results show that the general performance of experiential models （C correction model and method of moment）is better than physical models （Shepherd model and Sandmeier model） when high-accuracy DEM and atmospherieal parameters are lacking.

关 键 词：高分辨率影像 地形辐射校正 模型对比 C校正模型 矩匹配方法 Shepherd模型 Sandmeier模型 

分 类 号：P237[天文地球—摄影测量与遥感]