EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PROOF OF CONCEPT OF A LOW-COST TOOL FOR DATA COLLECTION AND TRAFFIC CONFLICT IDENTIFICATION AT CROSSWALKS

Samuel Seixas Martin, Fabiana Serra de Arruda

Revista Científica de Pesquisa Aplicada à Engenharia · 2026 · 8 (1) · 92-101

DOI: 10.29327/2214740.8.1-8 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Revista Científica de Pesquisa Aplicada à Engenharia, 2026. Verfasst von Samuel Seixas Martin und 1 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Urban traffic conflicts pose significant safety risks and present real-time monitoring challenges in increasingly complex environments. Addressing these issues, especially at pedestrian crossings, this study proposes an Artificial Intelligence-based video analysis tool that can be deployed immediately and at low cost for integrated data collection, as literature suggests. Using Python, our program applies object detection, filtering, and temporal tracking with physical marker correlation to count vehicles and pedestrians, measure their speeds, and categorize interactions into risk levels. In a real-world pilot on the University Campus, the tool accurately identified speed violations and high-risk crossing events. Its performance surpasses traditional manual data collection methods in both technical precision and economic efficiency, provided the camera has an unobstructed view. There are vast possibilities for this solution, which enables traffic planners to detect and mitigate conflicts with data-driven tools for the safer traffic that all human beings inherently deserve.

Zitieren

APA

Martin, S. S., & Arruda, F. S. D. (2026). PROOF OF CONCEPT OF A LOW-COST TOOL FOR DATA COLLECTION AND TRAFFIC CONFLICT IDENTIFICATION AT CROSSWALKS. Revista Científica de Pesquisa Aplicada à Engenharia, 8(1), 92-101. https://doi.org/10.29327/2214740.8.1-8

Vancouver

Martin SS, Arruda FSD. PROOF OF CONCEPT OF A LOW-COST TOOL FOR DATA COLLECTION AND TRAFFIC CONFLICT IDENTIFICATION AT CROSSWALKS. Revista Científica de Pesquisa Aplicada à Engenharia. 2026;8(1):92-101. doi:10.29327/2214740.8.1-8

BibTeX
@article{Martin2026,
  author = {Martin, Samuel Seixas and Arruda, Fabiana Serra de},
  title = {PROOF OF CONCEPT OF A LOW-COST TOOL FOR DATA COLLECTION AND TRAFFIC CONFLICT IDENTIFICATION AT CROSSWALKS},
  journal = {Revista Científica de Pesquisa Aplicada à Engenharia},
  year = {2026},
  volume = {8},
  number = {1},
  pages = {92--101},
  doi = {10.29327/2214740.8.1-8}
}
Kostenlos starten →