Using AI on data tied to places on the Earth, such as satellite and aerial images, maps, location traces from phones and vehicles, and sensor readings. It can map buildings and roads, track crops, spot flooding or deforestation, and watch for change over areas far too large for people to inspect by hand.
Technical view
Combines computer vision on remote sensing imagery, including multispectral and synthetic aperture radar data that sees through cloud, with spatial statistics and geographic information systems. Distinctive problems include very large images tiled into patches, varying resolution, revisit times and sensors, spatial autocorrelation that makes random train and test splits leak information, and labels that go out of date as the ground changes. Uses span agriculture, insurance, disaster response, urban planning, climate monitoring and defence intelligence, the last raising dual use and privacy questions.
Related
Part of AI Terminology Index, section: Modalities and Applications.