基于HJ-1 CCD数据的地表反照率反演

An algorithm for retrieving land surface albedo from HJ-1 CCD data

波段设置的有限性以及有限的观测角度限制了HJ-1 CCD数据在地表反照率反演方面的应用。该文利用高质量的POLDER-BRDF(双向反射分布函数)数据集分别在植被、裸土、冰雪地物覆盖下模拟了不同观测几何条件下CCD 4个波段的地表方向反射率的短波波段地表黑空反照率(直入扇出反照率)和白空反照率(扇入扇出反照率);基于太阳和传感器观测几何原理对模拟数据进行分格网最小二乘回归分析,构建了基于CCD 4个波段地表方向反射率和短波波段地表反照率估算模型;选取黑河高质量的HJ-1 CCD数据开展了短波波段地表黑、白空反照率的反演实验,并利用实测的地表反照率数据对模型的精度进行了评价。结果表明,在满足验证条件的23组数据中,有21组数据的绝对误差在0.03以内。

Because of the limitation of band and observation settings, it is difficult to use HJ - 1 CCD imagery to retrieve land surface albedo. In this paper, the authors developed a new algorithm for HJ - 1 CCD land surface albedo retrieval by introducing POLDER BRDF data as the support. In this algorithm, POLDER - BRDF datasets are used to simulate surface reflectance of HJ - 1 CCD four bands and short - wave white/black sky albedo under the conditions of various solar - view geometries and different land cover types （ vegetation, bare soil, snow and ice） ; then land surface albedo estimating model can be calculated under different grids using the Least Squares Fitting method. The input of this algorithm are four bands surface reflectance of HJ - 1 CCD camera. Twenty - three HJ- 1 CCD images were selected to retrieve land surface albedo using this proposed method. A comparison with measured surface albedo indicates that the error of 21/23 albedos is less than 0.03 in absolute value.