AI Powered Drowsiness and Lane Departure Warning System Using Deep Learning

Abstract

Driver drowsiness and unintended lane departure are two of the many factors that can increase the possibility of traffic accidents. The traditional driver assistance systems typically take one of the two approaches to detect the status of drivers and the road conditions; either monitor the driver or analyze the car positioning in the road. With the recent developments in the field of artificial intelligence and deep learning, it has become possible to analyze the face characteristics, eyes movement, yawning, road scene, and lane marks using camera-based system. In this regard, this review paper discusses an AI-driven driver safety system that involves the use of two cameras for the simultaneous detection of driver drowsiness and lane departure. The first camera is set toward the driver and analyses the face characteristics, eye closure, yawning, and other signs of fatigue through deep learning algorithms. The second camera is set toward the road and continuously analyzes the lane marks and position of the car to detect lane departure. The outputs of both the camera-based detection systems are processed at the decision layer and provide corresponding real-time warnings.