Deep learning enhanced road traffic analysis: Scalable vehicle detection and velocity estimation using PlanetScope imagery

This paper presents a method for detecting and estimating vehicle speeds using PlanetScope SuperDove satellite imagery, offering a scalable solution for global vehicle traffic monitoring. Conventional methods such as stationary sensors and mobile systems like UAVs are limited in coverage and constrained by high costs and legal restrictions. Satellite-based approaches provide broad spatial coverage, but face challenges, including high costs, low frame rates, and difficulty detecting small vehicles in medium-resolution imagery. We propose a Keypoint R-CNN model to track vehicle trajectories across RGB bands, leveraging band timing differences to estimate speed. Validation is performed using drone footage and GPS data that covers highways in Germany and Poland. Our model achieved a Mean Average Precision of 0.53 and velocity estimation errors of approximately 3.4 m/s compared to GPS data. The results of the drone comparison reveal underestimations, with average speeds of 112.85 km/h for satellite data versus 131.83 km/h from drone footage. While challenges remain with high-speed accuracy, this approach demonstrates the potential for scalable, daily traffic monitoring across vast areas, providing valuable insights into global traffic patterns. • Pose estimation deep learning model delineates moving vehicles from Planet imagery. • Trajectory can be detected as subtle signatures across satellite spectral bands. • An automated method uses RS imagery, offering global scalability and daily coverage. • A reproducible validation framework for estimating velocity by road segment. • The method enhances traffic data for large-scale coverage and continuous monitoring.