Road Surface Classification for Iran, Derived from OpenStreetMap and AI

Approximately 0.8963 million kilometers of roads in Iran are mapped, with AI-derived estimates classifying 16.1911% as paved and 7.6254% as unpaved. The dataset, created by the Heidelberg Institute for Geoinformation Technology (HeiGIT), combines OpenStreetMap data with deep learning predictions from Mapillary imagery and urban classification layers. It was last updated on March 17, 2026.

Use Cases

• Transportation planning based on paved and unpaved road classifications.
• Infrastructure analysis using road attributes like surface type, width, and lanes.
• Climate emissions assessment based on road surface data and urban/rural classification.
• GIS applications leveraging geospatial road network data with AI-derived surface labels.

Strengths

• Includes AI-derived surface classifications for over 0.8963 million kilometers of roads.
• Combines multiple data sources: OpenStreetMap, Mapillary imagery, and urban settlement layers.
• Provides explicit statistics, such as 0.1451 million km of paved roads and 0.6828 million km with missing surface data in OSM.

Limitations

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

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 the referenced arXiv papers.

Time Range

Mean timestamp of predictions (DL_mean_timestamp) is included, but overall temporal coverage is not explicitly stated.

Freshness

Last updated 2026-03 17.

Geography

Iran (Islamic Republic of).