The DeepFlood dataset provides high-resolution georeferenced images from both manned and unmanned aerial platforms, featuring detailed labels that go beyond simple binary distinctions. These labels include inundated vegetation, dry vegetation, open water, and others, making the dataset highly applicable for flood mapping across various landscapes. It uniquely incorporates SAR imagery alongside optical and UAV images, enabling a multi-modal approach to accurately delineate flooded areas.
Funding
NOAA Award NA21OAR4590358-T1-01:Rapid Floodwater Extent and Depth Measurements Using Optical UAV and SAR
NASA Award 80NSSC23M0051 : DEAP Institute : Harnessing Data Science for Flood Monitoring and Management