Development of a Python Algorithm for the Detection of Telephone Distractions in Drivers using YOLOv8 Neural Networks)

León León R.A.; Mendoza Gutierrez J.E.; Polo Del Rosario C.S. · Memorias de la Conferencia Iberoamericana de Complejidad, Informatica y Cibernetica, CICIC (2025) · DOI: 10.54808/cicic2025.01.341

Resumen

This article presents the development of an algorithm in Python using the YOLOv8 neural network to detect driver distractions due to the use of mobile phones. Knowing that distracted driving is a major cause of accidents, this project seeks to identify cell phone use in real time through videos captured by in-vehicle cameras. The system implementation includes the use of YOLOv8, a neural network architecture optimized for object detection. Resources such as the OpenCV library were used to process images and tools such as Pandas for the analysis of generated data. During testing, the model achieved an accuracy of 93% and demonstrated a real-time detection capability of 95%. However, there were limitations due to lighting factors and camera angle, which affected accuracy. The results suggest that the system has potential to improve road safety by alerting about distractions, although additional adjustments are required to optimize accuracy under different conditions. Copyright 2025. © por el International Institute of Informatics and Systemics.

Instituciones: Universidad Privada del Norte