Singapore Road Surface Classification from OSM and Deep Learning

Singapore's road network, with approximately 44,300 km of roads mapped in OpenStreetMap. The dataset, created by HeiGIT, classifies road surfaces as paved or unpaved using a hybrid deep learning approach on Mapillary imagery, augmented with urban classification layers. It was last updated on March 2, 2026.

Use Cases

• Transportation planning based on paved versus unpaved road classifications.
• Infrastructure analysis using predicted road length and surface type.
• Climate emissions assessment based on road surface material data.
• GIS applications integrating OSM attributes like highway type and urban classification.

Strengths

• Provides AI-derived estimates for 0.023 million km of paved roads and 0.0015 million km of unpaved roads.
• Augments OSM data with an additional 0.0021 million km of surface information from Mapillary imagery.
• Includes multiple classification schemes (OSM-priority and DL-priority) and urban/rural labels.

Limitations

• Approximately 44.797% (0.0198 million km) of road surface information is missing from the underlying OSM data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

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, with methodology detailed in referenced arXiv papers.

Time Range

Mean timestamp of predictions is provided via the DL_mean_timestamp field.

Freshness

Last updated 2026-03-02 17:18:30.244596; freshness should be verified.

Geography

Singapore.