Related Experiment Video
Updated: Aug 8, 2025
08:45
Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
Published on: October 24, 2012
14.7K
EEG-based driver states discrimination by noise fraction analysis and novel clustering algorithm
Rongrong Fu1, Zheyu Li1, Shiwei Wang2
1Department of Electrical Engineering, Measurement Technology and Instrumentation Key Lab of Hebei Province, Yanshan University, Qinhuangdao, China.
Biomedizinische Technik. Biomedical Engineering
|February 27, 2023
Summary
This study developed a noise fraction analysis method to remove electrooculography (EOG) artifacts from electroencephalogram (EEG) signals. This technique accurately identifies driver fatigue with over 90% accuracy.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Signal Processing
Background:
- Driver states significantly impact driving safety.
- Electroencephalogram (EEG) signals are crucial for distinguishing driver states.
- Noise and redundant information in EEG reduce signal quality and hinder accurate analysis.
Purpose of the Study:
- To propose an automated method for removing electrooculography (EOG) artifacts from EEG signals.
- To enhance the signal-to-noise ratio of EEG recordings for improved driver state analysis.
- To develop a robust algorithm for identifying denoised EEG signals and recognizing driver fatigue.
Main Methods:
- Collected multi-channel EEG data from drivers after prolonged driving and rest periods.
- Applied noise fraction analysis to separate EEG components and optimize the signal-to-noise quotient, effectively removing EOG artifacts.
- Utilized Fisher ratio space for representing denoised EEG characteristics and developed a novel cluster ensemble and probability mixture model (CEPM) for EEG identification.
Main Results:
- Noise artifacts were successfully removed from EEG signals using noise fraction analysis.
- The CEPM algorithm achieved high clustering accuracy, exceeding 90% for all participants.
- EEG mapping plots visually demonstrated the effectiveness of the denoising method.
Conclusions:
- Noise fraction analysis is an effective technique for EOG artifact removal in EEG signals.
- The developed CEPM algorithm accurately identifies denoised EEG signals, leading to high driver fatigue recognition rates.
- This method holds significant potential for improving driver safety systems.
