Wan Hua-Wei, Wang Jin-Di, Liang Shun-Lin, Qin Jun
School of Geography and Remote Sensing Science, Beijing Normal University, State Key Laboratory of Remote Sensing Science, Beijing 100875, China.
Guang Pu Xue Yu Guang Pu Fen Xi. 2009 Nov;29(11):3106-11.
Moderate-resolution imaging spectrometer (MODIS) and multi-angle imaging spectroradiometer (MISR) are two important sensors on TERRA satellite. The authors can have more spectral and multi-angular observations on the land surface objects by combining these two datasets. In the present paper, both MODIS and MISR observations were combined to estimate leaf area index (LAI) of land surface. The adjoining model and trust-region optimal algorithm were introduced into the framework of physical model inversion to speed up the running of the model inversion algorithm. And the algorithm allows the prior knowledge on the retrieved parameters to be input into the inversion procedure. The uncertainty and sensitivity matrix (USM) based analysis is helpful for selecting the observed data subset with more information and less noise to retrieve LAI. The measured LAI in situ and estimated LAI from ETM data were scaling-up to MODIS/MISR LAI product scale, and were taken as the ground truth to evaluate the new approach. The result suggests that combining two sensors datasets can improve the accuracy of LAI estimation, and the developed inversion method in this paper can be applied to the large area remote sensed image data effectively.
中分辨率成像光谱仪（MODIS）和多角度成像光谱辐射计（MISR）是TERRA卫星上的两个重要传感器。通过结合这两个数据集，作者能够对地表物体进行更多的光谱和多角度观测。在本文中，MODIS和MISR观测数据被结合起来以估算地表叶面积指数（LAI）。邻接模型和信赖域优化算法被引入到物理模型反演框架中，以加速模型反演算法的运行。并且该算法允许将检索参数的先验知识输入到反演过程中。基于不确定性和灵敏度矩阵（USM）的分析有助于选择具有更多信息且噪声更少的观测数据子集来检索LAI。将实地测量的LAI以及从ETM数据估算的LAI尺度放大到MODIS/MISR LAI产品尺度，并将其作为地面真值来评估新方法。结果表明，结合两个传感器数据集可以提高LAI估算的精度，并且本文所开发的反演方法能够有效地应用于大面积遥感影像数据。
