A Novel Fatigue Driving State Recognition and Warning Method Based on EEG and EOG Signals

Area of Science:

• Neuroscience and Biomedical Engineering
• Transportation Safety Systems

Background:

• Fatigue driving is a major cause of traffic accidents.
• Current fatigue detection methods often lack real-time performance and accuracy due to feature limitations and processing time.
• Existing methods using facial expressions, EEG, yawning, and PERCLOS are prone to errors and delays.

Purpose of the Study:

• To develop a real-time and accurate fatigue driving recognition system.
• To improve the feasibility and performance of fatigue detection for enhanced road safety.
• To integrate a fatigue detection system with IoT technology for early warnings and communication.

Main Methods:

• A novel Fast Support Vector Machine (FSVM) algorithm was developed.
• Electroencephalogram (EEG) and electrooculogram (EOG) signals were collected and preprocessed.
• Multiple features were extracted from EEG and EOG data, followed by FSVM classification for fatigue state recognition.

Main Results:

• The proposed FSVM algorithm demonstrated effective real-time recognition of driver fatigue.
• The system successfully integrated EEG and EOG data for accurate fatigue state identification.
• An Internet of Things (IoT)-based early warning system was designed, providing alerts to drivers and nearby vehicles.

Conclusions:

• The FSVM algorithm offers a feasible and real-time solution for detecting driver fatigue.
• The developed IoT-based system enhances traffic safety by providing timely warnings and communication.
• This approach significantly improves upon existing methods by ensuring accuracy and real-time performance in fatigue detection.