A Framework of Multi-Baseline PolInSAR Methods for Robust Forest Height Estimation

In this paper, the status of the existing multi-baseline polarimetric SAR interferometry (PolInSAR) methods is reviewed. Based on this, we propose another advanced strategy for robust forest height estimation using multi-baseline PolInSAR data. The new strategy focuses on dual-baseline combination selection from multi-baseline PolInSAR data, which ensures the robust performance of the forest height inversion by effectively improving the estimation of volume-only coherence. More specifically, the multi-baseline fusion can be performed based on, or not based on external data, depending on the availability of the auxiliary data to be used. We validate the proposed approach using airborne L-band and P-band SAR data acquired by the E-SAR and F-SAR system as well as LiDAR data during the BioSAR-2008 and AfriSAR-2016 campaign. The results show that the forest height maps derived by our MBPI strategy highly agrees with those derived by LiDAR, presenting an improvement compared with existing multi-baseline PolInSAR approaches, which could be considered as a promising tool for TanDEM-L or BIOMASS mission to invert vegetation parameters at a global scale.