Advancing Road Safety : A CNN-Bassed Approach to Drowsiness Detection with Semi-Supervised Learning

Abstract

Driver drowsiness detection is a critical area of research aimed at enhancing road safety and preventing accidents caused by fatigue. This study presents a novel approach for detecting driver drowsiness using a combination of semi-supervised learning and Convolutional Neural Networks (CNNs). The proposed method leverages the strength of CNNs in feature extraction and pattern recognition, coupled with semi-supervised learning to efficiently utilize both labeled and unlabeled data. By integrating these techniques, the system can effectively identify signs of drowsiness from video frames captured in real-time. The semi-supervised learning approach addresses the challenge of limited labeled datasets by incorporating a larger pool of unlabeled data to improve model robustness and accuracy. Extensive experiments demonstrate that our method achieves superior performance in drowsiness detection compared to traditional super- vised learning models, showing promise for real-world applications. The implementation of this system could significantly reduce the incidence of drowsiness-related accidents, contributing to safer driving environments.