A PRATICAL IMPLEMENATION OF AN INTELLIGENT DROWSY DRIVER MONITORING AND DETECTION MODEL

In the field of accident control due to driver's drowsiness, the conventional approach involved monitoring and analyzing the driver's physiological signals such as electrocardiography (ECG) and electroencephalography (EEG). However, these methods were complex and required high maintenance, prompting the need for alternative driver monitoring techniques. Driver drowsiness is a significant contributing factor to many vehicle accidents, impacting global human mortality rates. In this study, a convolutional neural network (CNN) was employed to develop a drowsy driver detection system. The dataset used for training the CNN model consisted of drowsy driver samples collected from the UCI repository. The trained model was then utilized to create the drowsy driver detection system. The classification model developed with CNN exhibited an impressive 99.7% accuracy in detecting drowsy behavior. Subsequently, the model was tested in practical experiments, where it successfully identified instances of drowsiness with high precision. This research highlights the effectiveness of using CNN for drowsy driver detection, providing a promising solution to mitigate the risks associated with driver drowsiness and reduce the occurrence of accidents caused by this factor.