Machine Learning in Model-Based Forest Height Inversion

Model-based (PM) forest height inversion from Polarimetric Interferometric Synthetic Aperture Radar (Pol-InSAR) measurements is today an established application demonstrated and validated at large scales for a wide variety of boreal and tropical forest sites at different frequencies (from X- down to P- band) ​[1], [2]​. Although the estimation performance obtained may depend on the individual observation spaces in each case, it is generally (very) convincing. However, as with any model-based inversion approach, there are inherent limitations that can restrict expected performance depending on the individual case ​[3]​