Dangerous Driving Detection Based on IoT

Abstract

Today, finding drowsy drivers is perhaps the most important step in stopping any traffic accident, everywhere in the globe. The objective of this study was to develop a smart alert approach for intelligent vehicles that can prevent drowsy driving impairment automatically. But being sleepy is a natural bodily occurrence that might occur for a variety of reasons. Therefore, it is necessary to create a reliable alert system to prevent the accident's cause. In this proposed research, we discuss a method for developing a drowsy driver alert system that analyses video stream processing (VSP) utilizing the eye blink concept employing the eye aspect ratio (EAR) and Euclidean distance of the eye. The face landmark algorithm is also employed as a reliable method for detecting eyes. The IoT module sends a warning message together with collision impact and location information when the driver's weariness is recognized, alerting with the help of a voice speaking over the Raspberry Pi monitoring system.