Road Surface Classification for Saint Helena from OSM and Satellite Imagery

A 2026 dataset from HeiGIT provides AI-derived road surface classifications for Saint Helena, distinguishing paved and unpaved roads. It is based on OpenStreetMap data augmented with deep learning predictions from Mapillary imagery and PlanetScope satellite data. The dataset includes approximately 0.0005 million kilometers of mapped roads, with surface information missing for about 75.728% of that length in the original OSM data.

Use Cases

• Transportation planning based on road surface type and predicted length.
• Infrastructure analysis using attributes like highway type, surface, and smoothness.
• Climate emissions assessment informed by paved versus unpaved road proportions.
• GIS applications leveraging urban classification attributes like rural/urban designation.

Strengths

• Includes AI-derived surface predictions (paved/unpaved) for roads where OSM data is missing.
• Combines data from multiple sources: OSM, Mapillary imagery, PlanetScope satellites, and urban settlement layers.
• Provides specific statistics: 0.0001 million km of paved roads (20.18%) and 0.0 million km of unpaved roads (4.092%).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic bias inherent to the source platforms and region.

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT)

Collection Method

Derived from OpenStreetMap and augmented using a hybrid deep learning approach on Mapillary and PlanetScope imagery.

Time Range

null

Freshness

Last updated 2026-03-02 17:04:18.099457; freshness should be verified.

Geography

Saint Helena