Ear-EEG Based-Driver Fatigue Detection System Augmented by Computer Vision

Driver fatigue is a significant danger to road safety, resulting in numerous accidents and fatalities on a global scale. To tackle this problem, researchers have been exploring innovative techniques for identifying and preventing driver drowsiness during real-world driving situations. This study proposes a novel approach for detecting driver fatigue by merging EEG (Electroencephalogram) data from sensors positioned behind the ear with computer vision-based analysis of facial characteristics. Behind-the-ear (BTE) EEG provides a more practical and user-friendly alternative than traditional scalp EEG methods. In addition to Ear-EEG signals, computer vision technology enhances fatigue detection accuracy by examining drivers’ facial images while driving. The study introduces a custom-designed wearable device for gathering EEG data from four sensor electrodes behind the ear. Continuous wavelet transform (CWT) converts these EEG signals into scalograms. These scalograms and facial images captured by a camera focused on key facial areas such as the left eye, right eye, mouth, and entire face serve as inputs for a deep learning model developed for identifying driver fatigue. Subsequently, a comparative assessment is conducted to gauge the performance of the proposed system when using only Ear-EEG signals, only camera images, or a combination of both data sources. The test results validate the practicality and effectiveness of the proposed system in identifying driver fatigue. Additionally, a companion smartphone application has been developed to simplify and promptly monitor and alert drivers when they exhibit drowsiness while driving in traffic.