基于统计方法的TanDEM-X林下地形反演

Estimation of subcanopy topography from TanDEM-X InSAR data based on statistical models

在森林区域,微波难以穿透森林达到地表,获取真实的地表高程,其获取的数字高程模型(DEM)包含植被信息。本文使用单基线单极化TanDEM-X数据和SRTM DEM,充分利用其统计信息,并以菲律宾试验区为例,成功将植被信号移除,得到林下地形。试验结果表明:林下地形反演精度从7.74 m提高到4.77 m,精度提高了38.3%,能够为平坦地区大面积林下地形反演提供技术支持。

In the forest area,it is difficult for microwaves to penetrate the forest to reach the surface to obtain the real surface elevation,and the obtained digital elevation model(DEM)contains vegetation information.In this paper,single-baseline single-polarization TanDEM-X data and SRTM DEM are used to make full use of their statistical information.Taking the Philippines test area as an example,the vegetation signal is successfully removed and obtained the understory topography.The experimental results show that the inversion accuracy of understory topography is increased from 7.74 m to 4.77 m,and the accuracy is increased by 38.3%,which can provide technical support for large-area understory topography inversion in flat areas.