Large-scale TanDEM-X InSAR sub-canopy topography inversion under insufficient observation information
The TerraSAR/TanDEM-X InSAR system is currently one of the most important global mapping satellite systems, the data acquired by this system is mainly single-baseline, single-polarized SAR data. Restricted by insufficient observation in-formation, vegetation, and surface scattering information cannot be directly separated from the data, which makes it hard to apply the system to the large-scale, high-precision inversion of terrain under forest. Therefore, this article proposes a method for large-scale terrain inversion under forest using single/dual baseline observations. The core concept of this method is as fol-lows: for overlapping zones (dual baseline observation), the RVoG model is used to estimate the terrain under forest and the forest height; for non-overlapping zones (single baseline observation), with the estimated tree height of the overlapping zone as the prior information, a simplified RVoG model (C-SINC) is used to estimate the terrain under forest. The new method was analyzed and verified in the Teruel experimental zone of Spain. Comparing the results with LiDAR DEM, the RMSE of sub-canopy topography acquired by this method is 5.98m, which is 20％ higher in precision than the RMSE of 7.43m obtained by InSAR DEM. For large-scale areas, the new method provides a reference for the application of TanDEM-X to estimate the sub-canopy Topography.
万方数据
维普
