Barbados Road Surface Classification from OpenStreetMap and AI Predictions

A 2026 dataset from the Heidelberg Institute for Geoinformation Technology (HeiGIT) provides AI-derived road surface classifications for Barbados. It contains approximately 4.7 kilometers of roads from OpenStreetMap, with deep learning predictions augmenting surface tags. The data is intended for transportation planning, infrastructure analysis, and GIS applications.

Use Cases

• Transportation planning based on paved versus unpaved road classifications.
• Infrastructure analysis using AI-predicted surface classes and OpenStreetMap attributes.
• Climate emissions assessment leveraging road surface type and urban/rural classification.
• GIS applications utilizing geospatial road network data with surface and highway tags.

Strengths

• Provides AI-derived surface classifications for approximately 4.7 km of roads.
• Includes multiple classification outputs (OSM, DL-prioritized, combined) for comparison.
• Augments OpenStreetMap data with predictions from Mapillary street-level imagery.
• Integrates urban classification layers (GHSU, AFRICAPOLIS) for contextual analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Approximately 3 km (64%) of road surface information is missing in the base OSM data.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT)

Collection Method

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

Time Range

The dataset reflects the state of OpenStreetMap and associated predictions as of its creation/update.

Freshness

Last updated 2026-03-02 17:18:04.618745; freshness should be verified.

Geography

Barbados