Hong Kong Road Surface, Width, and Passability Data (2020-2024)

This dataset maps 1,900 km of arterial roads in Hong Kong SAR using AI-derived attributes from 2020 and 2024 PlanetScope satellite imagery. Produced by HeiGIT, it provides surface type, width, and passability metrics to supplement OpenStreetMap data where surface tags are missing for 28% of the network.

Use Cases

• Monitoring infrastructure modernization by tracking 'Road Surface Change' from unpaved to paved between 2020 and 2024
• Route optimization for logistics using 'Road Width Class' to identify segments capable of supporting two-lane heavy truck flow
• Assessing regional connectivity through the 'Humanitarian Passability Index (HPI)' which scores logistical accessibility from 1 to 6

Strengths

• 89.2% accuracy for AI-derived road surface predictions
• Provides surface data for 28% of arterial roads that lack tags in OpenStreetMap
• Includes temporal change detection between 2020 and 2024 imagery

Limitations

• Geographic scope is limited strictly to the Hong Kong Special Administrative Region
• AI predictions may be affected by urban canyon effects or building shadows common in Hong Kong
• Restricted to high-capacity arterial roads only

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT)

Collection Method

Deep learning analysis of PlanetScope satellite imagery (3m resolution) integrated with OpenStreetMap road geometries.

Time Range

2020-2024

Freshness

Based on satellite imagery from 2020 and 2024; last updated in March 2026.

Geography

Hong Kong SAR, China