Post-Disaster Building Damage Detection from Earth Observation Imagery

DOI

The dataset consist of satellite and UAV imagery. The satellite imagery was derived from the xBD dataset (published by XView2 under the CC BY-NC-SA 4.0 license). The original images of size 1024x1024 were resized into different datasets of image sizes 32x32, 64x64 and 256x256. The UAV images were derived from a variety of UAV images, collected during various UAV-missions carried out in the last decade by ITC (University of Twente). The original images of varying sizes were resized into different datasets of image sizes 64x64, 256x256 and 512x512.A full description of the folder structure and content can be found in the README.

Post-Disaster Building Damage Detection from Earth Observation Imagery using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks

Date Submitted: 2021-05-10

Files not yet migrated to Data Station. For access to these files, please contact DANS at info@dans.knaw.nl.

Identifier
DOI https://doi.org/10.17026/dans-zsu-z2nc
Metadata Access https://phys-techsciences.datastations.nl/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.17026/dans-zsu-z2nc
Provenance
Creator S.M. Tilon ORCID logo
Publisher DANS Data Station Physical and Technical Sciences
Contributor M Th Koelen; F. Nex (Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente); N. Kerle (Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente); G Vosselman (Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente)
Publication Year 2021
Rights DANS Licence; info:eu-repo/semantics/openAccess; https://doi.org/10.17026/fp39-0x58
OpenAccess true
Contact M Th Koelen (Faculty of Geo-Information Science and Earth Observation)
Representation
Resource Type Dataset
Format application/zip
Size 19040
Version 1.0
Discipline Other