Hybrid AI-Physical Modeling for Penetration Bias Correction in X-band InSAR DEMs: A Greenland Case Study
This work addresses penetration bias correction in DEMs for glacial and snow-covered regions like Greenland, offering an incremental improvement by integrating existing physics with machine learning.
The paper tackled systematic elevation errors (penetration bias) in X-band InSAR DEMs over glacial regions by proposing a hybrid AI-physical modeling correction framework, which significantly reduced mean and standard deviation of DEM errors compared to a purely physical baseline and improved generalization over a pure ML approach in Greenland case studies.
Digital elevation models derived from Interferometric Synthetic Aperture Radar (InSAR) data over glacial and snow-covered regions often exhibit systematic elevation errors, commonly termed "penetration bias." We leverage existing physics-based models and propose an integrated correction framework that combines parametric physical modeling with machine learning. We evaluate the approach across three distinct training scenarios - each defined by a different set of acquisition parameters - to assess overall performance and the model's ability to generalize. Our experiments on Greenland's ice sheet using TanDEM-X data show that the proposed hybrid model corrections significantly reduce the mean and standard deviation of DEM errors compared to a purely physical modeling baseline. The hybrid framework also achieves significantly improved generalization than a pure ML approach when trained on data with limited diversity in acquisition parameters.