Buildings Damage Assessment

We focus, in this work, on buildings damage assessment (BDA). We propose to rely solely on RGB satellite images and follow a 2-stage deep learning-based approach, where first, buildings’ footprints are extracted using a semantic segmentation model, followed by classification of the cropped images into damaged or non-damaged building.

We conduct extensive experiments to select the best hyper-parameters, model architecture, and training paradigm, and we propose a new transfer learning-based approach that outperforms classical methods. Finally, we validate the proposed approach showing excellent accuracy and F1-score metrics.

Paper preprint can be fetched here.