Eritrea Road Surface Classification from OpenStreetMap and Deep Learning

Approximately 0.0262 million kilometers of roads are mapped in OpenStreetMap for Eritrea. The dataset classifies road surfaces as paved or unpaved using a hybrid deep learning approach based on Mapillary imagery, augmented with urban classification layers. It was published by the Heidelberg Institute for Geoinformation Technology (HeiGIT) and last updated in March 2026.

Use Cases

• Transportation planning based on the classification of paved versus unpaved roads.
• Infrastructure analysis based on road attributes like surface type, width, and lanes.
• Climate emissions assessment based on road surface data for vehicle emission modeling.
• GIS applications for routing and logistics based on the geospatial road network.

Strengths

• Provides AI-derived estimates for 0.0006 million km of paved and 0.0089 million km of unpaved roads.
• Integrates multiple data sources: OpenStreetMap, deep learning predictions from Mapillary, and urban classification layers.
• Includes detailed attributes such as predicted surface class, OSM tags, and urban/rural classification.

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.
• Approximately 0.0167 million km (63.5473%) of road surface information is missing in the underlying OSM data.

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.

Time Range

Mean timestamp of predictions is included in the 'DL_mean_timestamp' attribute.

Freshness

Last updated 2026-03 02 17:21:06.680168; freshness should be verified.

Geography

Eritrea.