Costa Rica Road Surface Classification from OSM and Deep Learning

A 2026 dataset from the Heidelberg Institute for Geoinformation Technology (HeiGIT) provides AI-derived road surface classifications for Costa Rica. It covers approximately 68,000 km of OpenStreetMap roads, with deep learning predictions augmenting surface data for paved and unpaved segments. The data is derived from OSM, Mapillary imagery, and urban layers, and is available in GeoJSON and GeoPackage formats.

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 modeling.
• GIS applications utilizing geospatial road networks with urban/rural classifications.

Strengths

• Provides AI-derived surface classification for approximately 68,000 km of roads.
• Augments OSM data with deep learning predictions from Mapillary imagery.
• Includes detailed OSM and urban classification attributes for analysis.

Limitations

• Approximately 61% of road surface information is missing in the underlying OSM data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

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

Collection Method

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

Time Range

null

Freshness

Last updated 2026-03 02 17:15:43.398417.

Geography

Costa Rica