Serbian Road Surface Classification with AI Predictions

Approximately 226,400 km of Serbian roads are mapped, with AI-derived estimates classifying 18.6% as paved and 16.2% as unpaved. This dataset, created by HeiGIT and updated in March 2026, combines OpenStreetMap data with deep learning predictions from Mapillary imagery to fill information gaps. It includes attributes for highway type, surface, and urban classification.

Use Cases

• Transportation planning based on paved and unpaved road classifications.
• Infrastructure analysis using attributes like highway type, surface, and width.
• Climate emissions assessment leveraging road surface data for vehicle emission models.
• GIS applications for routing and network analysis based on the geospatial road data.

Strengths

• Includes AI-derived surface predictions for over 226,000 km of roads.
• Combines OpenStreetMap data with deep learning predictions to address a 65.2% data gap in OSM.
• Provides multiple classification schemes (OSM-priority and DL-priority) for flexibility.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

Derived from OpenStreetMap and augmented with deep learning predictions based on Mapillary imagery, combined with GHSU and AFRICAPOLIS urban layers.

Time Range

null

Freshness

Last updated 2026-03-02 17:13:33.923835; freshness should be verified.

Geography

Serbia