Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Maciej Adamiak, Yulia Grinblat, Julian Psotta, Nir Fulman, Himshikhar Mazumdar, Shiyu Tang, Alexander Zipf

TL;DR

A Keypoint R-CNN model is proposed to track vehicle trajectories across RGB bands, leveraging band timing differences to estimate speed, demonstrating the potential for scalable, daily traffic monitoring across vast areas, and providing valuable insights into global traffic dynamics.

Abstract

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.

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

TL;DR

A Keypoint R-CNN model is proposed to track vehicle trajectories across RGB bands, leveraging band timing differences to estimate speed, demonstrating the potential for scalable, daily traffic monitoring across vast areas, and providing valuable insights into global traffic dynamics.

Abstract

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.

Paper Structure

This paper contains 26 sections, 7 equations, 10 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related works
  3. Materials and Methods
  4. Datasets
  5. PlanetScope Imagery
  6. Drone footage
  7. GPS tracks
  8. OpenStreetMap (OSM) road data
  9. Conceptual framework on ‘moving echoes’
  10. The proposed algorithm
  11. Detecting vehicle trajectories
  12. Keypoint correction
  13. Keypoint R-CNN pose estimation model
  14. Model validation metrics
  15. Velocity estimation
  16. ...and 11 more sections

Figures (10)

  • Figure 1: Drone footage vehicle tracking examples acquired on the A1 highway near Plichtow, Poland.
  • Figure 2: ‘Moving echoes’ on true colour SuperDove satellite imagery on the A7 highway; Hanover-Kassel, Germany.
  • Figure 3: An overview of the proposed algorithm for detecting moving vehicles and estimating their speed based on SuperDove images.
  • Figure 4: ‘Moving echoes’ with trajectory training labels.
  • Figure 5: Annotation of moving vehicles on Highway A7 and A2 (Germany) based on RGB images (a, b, c) of PlanetScope SuperDove scenes from 26.09.2023.
  • ...and 5 more figures