EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

An Intelligent Traffic Management System Using Computer Vision and Machine Learning

Sandeep Bindra, Chetna, Stuti, Shilpi

International Journal of Research and Review in Applied Science, Humanities, and Technology · 28.12.2024 · 263-270

DOI: 10.71143/72qgz725 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Research and Review in Applied Science, Humanities, and Technology, online erschienen am 28.12.2024. Verfasst von Sandeep Bindra und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fixed-time traffic signals cannot respond to hourly, directional and weather-driven changes in demand at urban intersections. This paper presents an intelligent traffic management system that combines computer vision and machine learning in one pipeline: background-subtraction vehicle detection with line-crossing counting, a Random Forest vehicle-type classifier, a machine-learning model that forecasts traffic flow 15 minutes ahead, and a vision-driven adaptive signal controller. The pipeline is evaluated by simulation on synthetic traffic video under five conditions, on a 60-day synthetic flow series, and on a four-approach signal-control model repeated over 30 random seeds. Detection F1 is 90.1% in normal conditions but falls to 43.7% in dense traffic; the classifier reaches 99.6% accuracy; the best flow predictor (SVR (RBF)) lowers RMSE by 27.6% against a naive baseline; and the adaptive controller reduces mean vehicle delay from 176.0 s to 27.2 s under unbalanced demand, while under balanced demand it matches or trails fixed time and Webster control. The results are from synthetic data and show where vision-based adaptive control helps, where it does not, and what must be done before field deployment.

Zitieren

APA

Bindra, S., Chetna, Stuti, & Shilpi (2024). An Intelligent Traffic Management System Using Computer Vision and Machine Learning. International Journal of Research and Review in Applied Science, Humanities, and Technology, 263-270. https://doi.org/10.71143/72qgz725

Vancouver

Bindra S, Chetna, Stuti, Shilpi. An Intelligent Traffic Management System Using Computer Vision and Machine Learning. International Journal of Research and Review in Applied Science, Humanities, and Technology. 2024:263-270. doi:10.71143/72qgz725

BibTeX
@article{Bindra2024,
  author = {Bindra, Sandeep and Chetna and Stuti and Shilpi},
  title = {An Intelligent Traffic Management System Using Computer Vision and Machine Learning},
  journal = {International Journal of Research and Review in Applied Science, Humanities, and Technology},
  year = {2024},
  pages = {263--270},
  doi = {10.71143/72qgz725}
}
Kostenlos starten →