Abstract:
The prevalence of accidents caused by driver drowsiness is a significant and pressing issue in
today's society. Despite the existence of various drowsiness detection systems, the high incidence
of such accidents indicates a need for more accurate and reliable solutions. This research aims to
address this problem by developing a Drowsiness Detection system using machine learning and
real-time image processing. The proposed system leverages public datasets containing images
and videos of drivers under various states of alertness. These datasets are preprocessed and fed
into a Convolutional Neural Network (CNN) model for training. The model is designed to detect
signs of drowsiness in real-time, providing timely alerts to potentially drowsy drivers. This
research represents a comprehensive effort to improve road safety by addressing the issue of
driver drowsiness. By utilizing advanced machine learning techniques and real-time image
processing, the proposed system aims to provide a more accurate and reliable solution to
drowsiness detection. The insights and developments from this research have the potential to
significantly reduce the number of accidents caused by driver drowsiness, thereby ensuring safer
roads and protecting the lives of drivers and pedestrians alike.
