Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

pygeovision: A Unified Open-Source Python Platform for Satellite Earth Observation AI

Domain:

geospatialenvironment and energy

Record type:

software
Creator:
App
Publisher:
Zenodo
Host:avatar
pygeovision is a unified open-source Python library for satellite Earth observation analysis and geospatial AI inference. Built on pygeofetch — a companion library abstracting satellite data access across 22+ providers — pygeovision covers the complete EO workflow: SAR and optical preprocessing, spectral index computation, cloud masking, change detection, flood mapping, SAR interferometry (InSAR), and inference via foundation models including Prithvi-EO-2.0 and DINOv3. A natural-language query interface (GeoAgent) allows users to describe an analysis in plain English. pygeovision is designed to make operational EO analysis accessible to researchers and government analysts in Africa and the Global South. If you use pygeovision in your research or operational work, please cite it using the metadata below. Citing the software helps the authors demonstrate research impact and continue development.

Visit

doi.org

Tasks

computer vision

Languages

Sar

Tags

earth observationremote sensingsynthetic aperture radarSARSentinel-1Sentinel-2Landsatsatellite imagerygeospatial analysisdeep learning+6

Licenses

MIT Licensehttps://opensource.org/licenses/MIT