| Summary: | Road safety remains a critical global issue, with driver distraction and drowsiness identified as leading causes of vehicle accidents. In Malaysia, about 532,125 road accidents were reported from January to October 2024, with 5364 fatal accidents. Hence, this study presents the development of a vehicle driver attention tracking system based on the analysis of brainwaves or Electroencephalogram (EEG) signals to enhance the safety driving while driving at the road. The proposed system utilizes the Unicorn Hybrid Black EEG device and LabVIEW software to monitor and classify driver attention states while driving vehicle. The attention states are Focus, Normal, and Drowsy. Raw EEG signals are preprocessed using band-pass filters to reduce noise and artifacts, followed by feature extraction technique to extract EEG features in term of mean, standard deviation, and spectral entropy. Then, the selected EEG features are fed to machine learning such as K-Nearest Neighbor (KNN) classifier where the classification accuracy exceeding 90 % for detecting driver attention levels during driving vehicle. This research combines advanced EEG signal processing and machine learning classification to create a promising approach to reduce the likelihood of accident caused by lack of attention or drowsiness during driving vehicle.
|
|---|