Design of Emergency Alarm System for Drowsiness Detection Using YOLO Method Based on Raspberry Pi
Year: 2024
Type: article
Abstract: Drowsiness is one of the main factors causing traffic accidents that often lead to fatalities, as drowsy drivers lose concentration. Therefore, drowsiness detection in car drivers is very important to prevent accidents. In this research, an emergency alarm system for drowsiness detection using YOLO method based on Mini PC is designed. This drowsiness detection system uses a camera to take pictures of the driver's face and the YOLO algorithm to detect the eyes. If the driver's eyes are detected to be closed, the system will give a warning in the form of a buzzer, display on the LCD, and water spray to wake up the driver. Test results show that the system has an average accuracy rate of 88% under optimal lighting conditions and 66.8% under low lighting conditions. The system also records detection and response times for further analysis, demonstrating the ability to not only detect the driver's condition but also record the time of the event. The data results show that the system is capable of detecting “Awake”, “Drowsy”, and “Asleep” states with an accuracy rate of 80%, as well as providing effective warnings to the driver when it detects signs of drowsiness.
Authors: Dinda Ayu Permatasari, Gillang Al Azhar, Muhammad Rifqi Zharfan, Anindya Dwi Risdhayanti, Arief Rahman Hidayat, Denda Dewatama
Institution: Politeknik Negeri Malang
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4408467472