Drowsy Driver Detection using Eye-Tracking through Machine Learning

• Published in: 2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC) pp. 1916 - 1923
• Main Authors: S, Akshay, MB, Abhishek, D, Sudhanshu, C, Anuvaishnav
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.08.2021
• Subjects: Adaptive Boosting; Cameras; Communication systems; DLib; Driver Assistance; Driver Drowsiness; Driver Monitoring System; Eye Tracking; Feeds; Gaze tracking; Haar Cascade Classifier; HOG; Machine learning; Organizations; PERCLOS; Road accidents; SVM
• DOI: 10.1109/ICESC51422.2021.9532928

Eye tracking is one of the most useful but underutilized technologies in today's world. It can be used in a variety of ways now that the technology is available. We propose an implementation that has not been done but should be after studying several ways to process the same. In the field of Advanced Driving Assistance Systems, tracking drivers' eyes is a hot topic (ADAS). According to data from the World Health Organization (WHO), approximately 1-1.25 million people die and 20-50 million people suffer from non-fatal injuries in road accidents each year around the world. And a high majority of these collisions are caused by drowsy driving. Our paper explores a possible implementation that could help detect drowsiness given a subject, a phone camera and a single board computer. We establish a connection between the phone and the system using a network connection that streams the camera feed onto the system, which further performs computations to determine the drowsiness of the driver.