DROWSINESS DETECTION USING YOLOV8

Abstract

Because of their unique driving characteristics, car drivers have got particular driving techniques, knowledge and attitudes. Driver tiredness has been a severe issue affecting road safety; consequently, it is critical to create an effective drowsiness detection algorithm to avoid road accidents. Various research efforts have been directed toward the challenge of detecting abnormal human driver behaviors in order to study the frontal face of the drivers and automotive dynamics using computer vision techniques. However, traditional approaches cannot capture complex the driver behavior aspects. However, since the inception of deep learning architectures, a large number of researches have been conducted to assess and recognize driver drowsiness using neural network techniques. This project:” Drowsiness Detection using Yolov8” is based on the Object Detection by Ultralytics Yolov8 and used to classify, analyze and detect the drowsiness of the users. The purpose of this project is to help minimize traffic accidents due to drowsiness or alert office worker to continue their work properly. The final experimental findings reveal that the algorithm described in this paper can accurately recognize proper and wrong facial expressions with a precision of 96.9%.