HJ-1 A/B CCD的陆地大气校正——具有地理信息的地表反射率产品(英文)

This paper proposed a method to retrieve the land surface reflectance from the HJ-1A/B CCD data. The aerosol optical depth(AOD), the most important factor affecting the atmospheric correction of CCD images at all bands, is proposed to retrieve from the CCD imagery by the approach of dense dark vegetation(DDV) method. A look-up table in terms of the transmittances, the path radiances and the atmospheric spherical albedo as functions of the AOD was established for a variety of sun-sensor geometry and aerosol loadings. The atmospheric correction is then achieved with the look-up table and the MODIS surface reflectance output(MOD09) as the priori datasets. Based on the retrieved AOD and the look-up table of atmospheric correction coefficients, the land surface reflectance was retrieved for the HJ-1A/B data according to the atmospheric radiative transfer equation. Some in-situ measurement Data for Yanzhou of Shandong province in East China and MODIS land surface reflectance products MOD09 are used to preliminarily validate the proposed method. The results show that the proposed method can remove effectively the atmospheric contributions, and the overall accuracy of the retrieval land surface reflectance can be improved substantially.     

HJ-1 CCD数据大气校正方法研究

本文开展了HJ-1CCD相机数据大气校正方法的研究工作。基于辐射传输模型构建了不同大气条件下的大气校正系数查找表;大气校正中用到的气溶胶光学厚度数据基于浓密植被区域红蓝波段地表反射率之间的关系反演得到。与对应当天的MODIS地表反射率数据的对比分析表明,本文提出的大气校正方法具有较高的精度。本文还从气溶胶光学厚度的反演精度、大气水汽含量的变化、辐射定标精度、海拔高度等方面对大气校正的不确定性进行了分析。