IoT-Enabled Drowsiness Detection Systems for Enhanced Road Safety Across Diverse Vehicle Types

This paper proposes the development of multiple driver drowsiness detection systems to enhance road safety across different types of vehicles, including two/three-wheelers, cars, and commercial trucks. The research encompasses three primary components: an initial prototype, a wearable device model, and a webcam-based model. The prototype utilizes infrared (IR) sensors to detect driver blinking patterns, activating an alarm and gradually stopping the vehicle when drowsiness is detected. The wearable device model effectively integrates IR sensors into helmets and glasses, making it a cost-effective and efficient solution for two/three-wheelers, which are prone to higher accident rates. The webcam-based model employs computer vision and machine learning techniques, using parameters such as Eye Aspect Ratio (EAR), Mouth Aspect Ratio (MAR), and excessive yawning to detect drowsiness. If drowsiness is detected, the system sends live location messages to emergency contacts. This model also features app support for real-time analysis of driver performance, benefiting both personal use and commercial monitoring of truck operators transporting hazardous goods. Our findings suggest that these models provide a versatile and impactful approach to reducing road traffic accidents.