Towards a Symbiosis of Model-Based and Machine Learning Forest Height Estimation based on TanDEM-X InSAR

Abstract

Recently, the German TanDEM-X mission provided valuable acquisitions for developing models for the single-pass single-pol X-band forest height inversion. In this paper, the assessment of the two machine learning approaches to estimate forest height from the interferometric coherence are investigated and compared to the state-of-art physical models over Gabon. The contribution of this work is toward the analysis of two approaches: Approach 1 is an implementation of a conventional ML approach. Approach 2 is the first attempt to integrate model-based knowledge in the ML approach and use a single input variable.