Road Surface Classification for Martinique from OSM and Satellite Imagery

Martinique road surface data, distinguishing paved and unpaved roads, derived from OpenStreetMap and augmented with deep learning predictions from Mapillary imagery. The dataset includes AI-derived classifications and OSM attributes, created by the Heidelberg Institute for Geoinformation Technology (HeiGIT) and last updated in March 2026. It is intended for transportation planning, infrastructure analysis, and GIS applications.

Use Cases

• Transportation planning based on road surface type classifications.
• Infrastructure analysis using attributes like highway type, surface, and width.
• Climate emissions assessment informed by road network composition.
• GIS applications leveraging geospatial road network data.

Strengths

• Includes AI-derived surface classifications (paved/unpaved) from a hybrid deep learning approach.
• Combines OpenStreetMap data with predictions from Mapillary street-level imagery.
• Enriched with urban classification attributes from GHSU and AFRICAPOLIS layers.
• Available in geospatial formats (GEOJSON, GEOPACKAGE).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• 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 (GHSU 2019, AFRICAPOLIS 2020).

Collection Method

Surface classification derived using a hybrid deep learning approach; data originates from OSM.

Time Range

The mean timestamp of predictions used is recorded in the 'DL_mean_timestamp' field.

Freshness

Last updated 2026-03-02 17:05:53.903197; freshness should be verified.

Geography

Martinique.