Curating flood extent data and leveraging citizen science for benchmarking machine learning solutions

Image | Flood | Preparedness

This is an image dataset for flood extent detection. It is a 4.11 GB download and is introdcued for image segmentation.

  • ML task type: Image segmentation
  • Data Source: Earth Observation Data and GeoSpatial Imagery (SAR)
  • Size: 4.11 GB
  • Timespan: N/A
  • Geographical Coverage: Bangladesh, Florence, Nebraska, North Alabama, Red River North
  • Baseline Information
  • Evaluated on: UNet, Feature Pyramid Network(FPN)
  • Metrics used: Intersection over Union
  • Results as reported in original paper: IOU: 0.6198 (UNet), 0.6021 (FPN)
  • Find out more by:
  • Reading the research paper
  • Note: This version of information is before the paper has been peer reviewed. At the point of data extraction, it is under consideration at Earth and Space Science.

Shubhankar Gahlot, Muthukumaran Ramasubramanian, Iksha Gurung, Ronny Hansch, Andrew Molthan, and Manil Maskey. Curating flood extent data and leveraging citizen science for benchmarking machine learning solutions. Earth and Space Science Open Archive, page 9, 2022.