Loading...
Loading...
Available on 1 platform
Sign in to view source links and access this dataset
Delivering AI-derived road surface, width, and passability attributes for approximately 5,700 km of arterial roads in Mauritania. Produced by the Heidelberg Institute for Geoinformation Technology (HeiGIT), the data utilizes PlanetScope satellite imagery from 2020 and 2024 to enhance OpenStreetMap (OSM) records. It specifically targets motorway, trunk, primary, and secondary road classes which form the country's national connectivity backbone.
Data is provided in GeoJSON and GeoPackage formats. Users should consult the associated ArXiv paper (2512.04092) for detailed methodology on the deep learning models used for surface and width prediction.