Benchmark Dataset for Automatic Damaged Building Detection from Post-Hurricane Remotely Sensed Imagery

Multimodal (Vector data and Image) | Hurricane | Response

FEMA and NOAA is a dataset for hurricane damaged building detection. It contains vector (FEMA) and image (NOAA) data and is introduced for image segmentation and multiclass (ordinal) classification.

  • ML task type: Image segmentation, multiclass (ordinal) classification
  • Data Source: Official Data (national); Earth Observation Data and GeoSpatial Imagery
  • Size: Around 156,099 damage assessments (FEMA original data); 400 GB Images (NOAA original data)
  • Timespan: 2017
  • Geographical Coverage: Greater Houston Area
  • Baseline Information
  • Evaluated on: unreported
  • Metrics used: unreported
  • Results as reported in original paper: unreported

Sean Andrew Chen, Andrew Escay, Christopher Haberland, Tessa Schneider, Valentina Staneva, and Youngjun Choe. Benchmark dataset for automatic damaged building detection from post-hurricane remotely sensed imagery. arXiv preprint arXiv:1812.05581, 2018.