Egypt Road Surface and Passability: 49,500 km of AI-Derived Metrics

HeiGIT generated this dataset of AI-derived road surface, width, and passability metrics for 49,500 km of arterial roads in Egypt. It utilizes PlanetScope satellite imagery from 2020 and 2024 to classify motorways, trunk, primary, and secondary road segments. The analysis provides surface information for 47,300 km of roads that previously lacked surface tags in OpenStreetMap.

Use Cases

• Planning humanitarian logistics by filtering for high-passability routes using the 'Humanitarian Passability Index (HPI)'
• Identifying infrastructure upgrades by detecting transitions from unpaved to paved in the 'Road Surface Change' field
• Determining vehicle size constraints for transport routes using 'Road Width Class' to identify segments narrower than 3.5m

Strengths

• 89.2% accuracy for surface predictions compared to 64.7% for OSM tags
• Covers 47,300 km of roads missing OSM surface tags
• Includes multi-temporal (2020 vs 2024) change detection

Limitations

• Geographic scope is limited to Egypt
• AI-derived width classes may have resolution-based margins of error
• Excludes minor road networks

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

Deep learning analysis of PlanetScope satellite imagery combined with OpenStreetMap road geometries

Time Range

2020 - 2024

Freshness

Last updated March 2026; contains imagery-derived data from 2020 and 2024.

Geography

Egypt