The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B3-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B3-2021, 741–748, 2021
https://doi.org/10.5194/isprs-archives-XLIII-B3-2021-741-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B3-2021, 741–748, 2021
https://doi.org/10.5194/isprs-archives-XLIII-B3-2021-741-2021

  29 Jun 2021

29 Jun 2021

AUTOMATED BUILDING SEGMENTATION AND DAMAGE ASSESSMENT FROM SATELLITE IMAGES FOR DISASTER RELIEF

X. Yuan1, S. M. Azimi1, C. Henry1, V. Gstaiger1, M. Codastefano3, M. Manalili3, S. Cairo3, S. Modugno3, M. Wieland2, A. Schneibel2, and N. Merkle1 X. Yuan et al.
  • 1Remote Sensing Technology Institute, German Aerospace Center (DLR), Oberpfaffenhofen, Germany
  • 2German Remote Sensing Data Center, German Aerospace Center (DLR), Oberpfaffenhofen, Germany
  • 3UN World Food Programme, Rome, Italy

Keywords: Satellite Imagery, Damage Assessment, Deep Learning, Building Segmentation, Crisis Management

Abstract. After a natural disaster or humanitarian crisis, rescue forces and relief organisations are dependent on fast, area-wide and accurate information on the damage caused to infrastructure and the situation on the ground. This study focuses on the assessment of building damage levels on optical satellite imagery with a two-step ensemble model performing building segmentation and damage classification trained on a public dataset. We provide an extensive generalization study on pre- and post-disaster data from the passage of the cyclone Idai over Beira, Mozambique, in 2019 and the explosion in Beirut, Lebanon, in 2020. Critical challenges are addressed, including the detection of clustered buildings with uncommon visual appearances, the classification of damage levels by both humans and deep learning models, and the impact of varying imagery acquisition conditions. We show promising building damage assessment results and highlight the strong performance impact of data pre-processing on the generalization capability of deep convolutional models.