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 high-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 covering 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. Results from 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 dynamics.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Deep Learning Enhanced Road Traffic Analysis: Scalable Vehicle Detection and Velocity Estimation Using PlanetScope Imagery


    Beteiligte:
    Adamiak, Maciej (Autor:in) / Grinblat, Yulia (Autor:in) / Psotta, Julian (Autor:in) / Fulman, Nir (Autor:in) / Mazumdar, Himshikhar (Autor:in) / Tang, Shiyu (Autor:in) / Zipf, Alexander (Autor:in)

    Erscheinungsdatum :

    2024


    Medientyp :

    Preprint


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Vehicle Traffic Estimation Using Deep Learning

    Meetkumar Patel / Daniel Silver | DOAJ | 2022

    Freier Zugriff

    Deep Learning based Vehicle Detection in Aerial Imagery

    Sommer, Lars Wilko | TIBKAT | 2022

    Freier Zugriff

    Enhanced Deep Learning Model for Road Transportation Safety with Accident Detection and Traffic Flow Analysis

    Santosh Kumar Patra, P. / Hari Krishna, B. / Krishna, Janapati Chaitanya et al. | Springer Verlag | 2024


    Deep Learning based Vehicle Detection in Aerial Imagery

    Sommer, Lars Wilko | GWLB - Gottfried Wilhelm Leibniz Bibliothek | 2022

    Freier Zugriff

    Deep learning based vehicle detection in aerial imagery

    Sommer, Lars Wilko | TIBKAT | 2021

    Freier Zugriff