Driver’s Drowsiness Detection System for Accident Prevention

Computer Engineering

Over the years, driver fatigue has been directly linked to the majority of accidents. In several industries, automation has played a crucial role in promoting uniformity and raising user quality of life. Although many drowsiness detection systems were created over the past ten years based on a variety of factors, the systems still needed to be improved in terms of effectiveness, accuracy, cost, speed, and availability, among other things. An integrated strategy is suggested in this study and is dependent on the PERCLOS (Eye and Mouth Closure Status) as well as the computation of the new proposed vector FAR (Facial Aspect Ratio), which is comparable to EAR and MAR. This aids in identifying the status of closed eyes or an opened mouth, such as when yawning, as well as any frames with hand motions like nodding or covering an open mouth with the hand, which are instinctive human behaviours used to manage tiredness. The system also integrated techniques and gradient patterns based on textural information to locate the driver's face in different orientations and to recognise sunglasses on the driver's face. Scenarios like hands covering the driver's eyes or lips while nodding or yawning were also identified and treated. The suggested work demonstrated greater accuracy when tested on datasets including NTHU-DDD, YawDD, and a proposed dataset EMOCDS (Eye and Mouth Open Close Data Set), and it delivers findings generally by taking numerous factors into account

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.