EEG ANALYSIS OF SLEEPY DRIVERS: IDENTIFYING KEY BRAINWAVE INDICATORS OF FATIGUE
Loading...
Files
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Malaysia Sarawak (UNIMAS)
Abstract
Description
Driver fatigue is considered a major cause of traffic accidents, injuring millions and killing many each year. The detection systems using existing methodologies based on external indicators, such as the movement of eyes or the pattern of the vehicle, are not very reliable and have low early detection capability. The present research investigates electroencephalogram (EEG) signals to identify a pattern of brainwaves that indicates fatigue, thereby improving realtime driver drowsiness detection systems. This study aims to apply machine learning techniques comprising Convolutional Neural Networks (CNN), Random Forest, and Support Vector Machines (SVM) to classify EEG data from alert and fatigued drivers; find important features in EEG signals related to fatigue that could be integrated into real-time driver monitoring system and test the accuracy of the developed models in detecting driver fatigue states. This study utilizes an open-access EEG dataset collected under controlled conditions, concentrating on the Delta, Theta, Alpha, and Beta frequency bands. Systematic methodology includes preprocessing, feature extraction, and classification modeling employed to analyze data. It is anticipated that specific EEG patterns, particularly those within the Theta and Alpha bands, will demonstrate a significant correlation with driver fatigue. The results are expected to provide the basis for developing robust and efficient drowsiness detection systems that will enhance road safety and reduce fatigue-related accidents. This research is important for the development of computer science, the improvement of transport safety, and the enrichment of human-machine interaction knowledge.