The multimedia satellite task at mediaeval 2018: emergency response for flooding events

Image | Flood | Preparedness

Flood Classification from Social Multimedia (FCSM) and Flood Detection from Satellite Imagery (FDSI) are datasets for road passability assessment, flood detection and flood classification. FCSM contains 7,387 tweets (development set), 3,683 images and features (test set) and FDSI contains 1,438 image patches (development set), 226 image patches (test set). They are both introduced for multilabel classification.

  • ML task type: Multilabel classification
  • Data Source: Social Media (Twitter); Earth Observation Data and GeoSpatial Imagery (Satellite)
  • Size: FCSM: 7,387 Tweets (Development set), 3,683 Images and features (Test set); FDSI: 1,438 Image patches (Development set), 226 Image patches (Test set)
  • Timespan: 2017
  • Geographical Coverage: Social Media: Areas affected by Hurricanes Harvey, Irma, Maria; Satellite: Houston Area
  • Baseline Information
  • Evaluated on: unreported
  • Metrics used: macro averaged F1-Score (Flood classification), metric F1-Score (Flood detection)
  • Results as reported in original paper: unreported

Benjamin Bischke, Patrick Helber, Zhengyu Zhao, Jens de Bruijn, and Damian Borth. The multimedia satellite task at mediaeval 2018 emergency response for flooding events. 10 2018.