Georgia Road Surface Classification from OSM and AI Predictions

Approximately 0.1136 million kilometers of roads from OpenStreetMap in Georgia are classified as paved or unpaved using a hybrid deep learning approach. The dataset, created by HeiGIT and last updated in March 2026, provides AI-derived estimates that 12.9% of roads are paved and 19.8% are unpaved, while 67.3% of surface information is missing from OSM. It is augmented with predictions from Mapillary imagery and urban classification layers.

Use Cases

• Transportation planning based on road surface type and length estimates.
• Infrastructure analysis using attributes like highway type, surface, and smoothness.
• Climate emissions assessment leveraging the distinction between paved and unpaved roads.
• GIS applications utilizing the geospatial road network and urban/rural classification.

Strengths

• Provides AI-derived surface classification for approximately 0.1136 million km of roads.
• Includes multiple classification schemes (OSM-priority and DL-priority) for comparison.
• Augments OSM data with deep learning predictions from Mapillary imagery.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Approximately 67.3% of road surface information is reported as missing from the underlying OSM data.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

OpenStreetMap (OSM), augmented with deep learning predictions from Mapillary imagery and urban layers (GHSL, AFRICAPOLIS).

Collection Method

Surface classification derived using a hybrid deep learning approach, as detailed in referenced papers.

Time Range

Mean timestamp of predictions is provided in the 'DL_mean_timestamp' field.

Freshness

Last updated 2026-03-02 17:23:33.429471; freshness should be verified.

Geography

Georgia (country).