Belgium Road Surface Classification from OpenStreetMap and AI Predictions

Road surface data for Belgium, distinguishing paved and unpaved roads, derived from OpenStreetMap and augmented with deep learning predictions from Mapillary imagery. The dataset covers approximately 0.2481 million kilometers of roads, with AI-derived estimates classifying 38.3796% as paved and 13.2798% as unpaved. It was published by the Heidelberg Institute for Geoinformation Technology (HeiGIT) and last updated in March 2026.

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 based on paved vs. unpaved road distribution.
• GIS applications leveraging geospatial road network data and urban classification attributes.

Strengths

• Covers approximately 0.2481 million kilometers of roads in Belgium.
• Provides AI-derived surface estimates for roads where OSM data is missing, adding classification for 4.7806% of the missing information.
• Includes multiple classification schemes (OSM-priority and DL-priority) and urban/rural labels.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

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

Collection Method

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

Time Range

null

Freshness

Last updated 2026-03-02 17:24:41.342999; freshness should be verified.

Geography

Belgium