DROWSY DRIVER DETECTION USING HEAD POSTURE: A DEEP LEARNING APPROACH
Abstract
ABSRACT
Road accidents due to driver drowsiness are a major safety issue, leading to many deaths and injuries globally. Drowsiness affects a driver's alertness, reaction time, and decision-making, often resulting in serious accidents. Current methods for detecting drowsiness, like self-reporting or physiological sensors, are not practical or reliable for everyday use. There is a need for a non-intrusive, real-time monitoring system to identify drowsiness before accidents happen. A new system was introduced in this study. It uses DLIB’s facial landmark predictor for head posture and a ResNet-50 classifier to identify drowsy drivers accurately by assessing facial features like head tilt and eye closure, proving effective in different lighting and angles. Experimental result showed accuracy of 90%, precision of 0.88, recall of 0.88 and fl-score of 0.88.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.