Road Network and Travel Time Extraction from Multiple Look Angles with SpaceNet Data
arXiv:2001.05923 · doi:10.1109/IGARSS39084.2020.9324091
Abstract
Identification of road networks and optimal routes directly from remote sensing is of critical importance to a broad array of humanitarian and commercial applications. Yet while identification of road pixels has been attempted before, estimation of route travel times from overhead imagery remains a novel problem, particularly for off-nadir overhead imagery. To this end, we extract road networks with travel time estimates from the SpaceNet MVOI dataset. Utilizing the CRESIv2 framework, we demonstrate the ability to extract road networks in various observation angles and quantify performance at 27 unique nadir angles with the graph-theoretic APLS_length and APLS_time metrics. A minimal gap of 0.03 between APLS_length and APLS_time scores indicates that our approach yields speed limits and travel times with very high fidelity. We also explore the utility of incorporating all available angles during model training, and find a peak score of APLS_time = 0.56. The combined model exhibits greatly improved robustness over angle-specific models, despite the very different appearance of road networks at extremely oblique off-nadir angles versus images captured from directly overhead.
arXiv comment4 pages, 5 figures. To appear at the 2020 IEEE International Geoscience and Remote Sensing Symposium
References in corpus (9)
- Deep learning in agriculture: A survey
- Deep learning in remote sensing: a review
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- SpaceNet: A Remote Sensing Dataset and Challenge Series
- A Large Dataset of Object Scans
- SpaceNet MVOI: a Multi-View Overhead Imagery Dataset
- City-Scale Road Extraction from Satellite Imagery v2: Road Speeds and Travel Times
- From Satellite Imagery to Disaster Insights
- Benchmark Dataset for Automatic Damaged Building Detection from Post-Hurricane Remotely Sensed Imagery