A Review of EEG Signal Features and their Application in Driver Drowsiness Detection Systems

Detecting drowsiness in drivers, especially multi-level drowsiness, is a difficult problem that is often approached using neurophysiological signals as the basis for building a reliable system. In this context, electroencephalogram (EEG) signals are the most important source of data to achieve successful detection. In this paper, we first review EEG signal features used in the literature for a variety of tasks, then we focus on reviewing the applications of EEG features and deep learning approaches in driver drowsiness detection, and finally we discuss the open challenges and opportunities in improving driver drowsiness detection based on EEG. We show that the number of studies on driver drowsiness detection systems has increased in recent years and that future systems need to consider the wide variety of EEG signal features and deep learning approaches to increase the accuracy of detection.

Download Full-text

EEG signal classification for drowsiness detection using wavelet transform and support vector machine

IAES International Journal of Artificial Intelligence (IJ-AI) ◽

10.11591/ijai.v10.i2.pp501-509 ◽

2021 ◽

Vol 10 (2) ◽

pp. 501

Author(s):

Novie Theresia Br. Pasaribu ◽

Timotius Halim ◽

Ratnadewi Ratnadewi ◽

Agus Prijono

Keyword(s):

Support Vector Machine ◽

Wavelet Transform ◽

Support Vector ◽

Eeg Signal ◽

Drowsiness Detection ◽

Driver Drowsiness ◽

Electroencephalogram Eeg ◽

Behavioral Methods ◽

Physiological Method ◽

Fold Cross Validation

<span id="docs-internal-guid-ed628156-7fff-8934-2369-94f011b043ca"><span>There are several categories to detect and measure driver drowsiness such as physiological methods, subjective methods and behavioral methods. The most objective method for drowsiness detection is the physiological method. One of the physiological methods used is an electroencephalogram (EEG). In this research wavelet transform is used as a feature extraction and using support vector machine (SVM) as a classifier. We proposed an experiment of retrieval data which is designed by using modified-EAR and EEG signal. From the SVM training process, with the 5-fold cross validation, Quadratic kernel has the highest accuracy 84.5% then others. In testing Driving-2 process 7 respondents were detected as drowsiness class, and 3 respondents were detected as awake class. In the testing of Driving-3 process, 6 respondents were detected as drowsiness class, and 4 respondents were detected as awake class. </span></span>

Download Full-text

EEG-based System Using Deep Learning and Attention Mechanism for Driver Drowsiness Detection

10.1109/ivworkshops54471.2021.9669234 ◽

2021 ◽

Author(s):

Miankuan Zhu ◽

Haobo Li ◽

Jiangfan Chen ◽

Mitsuhiro Kamezaki ◽

Zutao Zhang ◽

...

Keyword(s):

Deep Learning ◽

Attention Mechanism ◽

Drowsiness Detection ◽

Driver Drowsiness

Download Full-text

Driver Drowsiness Detection Using Deep Learning

Advances in Intelligent Systems and Computing - Applied Information Processing Systems ◽

10.1007/978-981-16-2008-9_7 ◽

2021 ◽

pp. 73-82

Author(s):

Ajinkya Rajkar ◽

Nilima Kulkarni ◽

Aniket Raut

Keyword(s):

Deep Learning ◽

Drowsiness Detection ◽

Driver Drowsiness

Download Full-text

Bias Remediation in Driver Drowsiness Detection Systems Using Generative Adversarial Networks

IEEE Access ◽

10.1109/access.2020.2981912 ◽

2020 ◽

Vol 8 ◽

pp. 55592-55601 ◽

Cited By ~ 1

Author(s):

Mkhuseli Ngxande ◽

Jules-Raymond Tapamo ◽

Michael Burke

Keyword(s):

Generative Adversarial Networks ◽

Drowsiness Detection ◽

Detection Systems ◽

Driver Drowsiness ◽

Adversarial Networks

Download Full-text

Methodology and initial analysis results for development of non-invasive and hybrid driver drowsiness detection systems

The 2nd International Conference on Wireless Broadband and Ultra Wideband Communications (AusWireless 2007) ◽

10.1109/auswireless.2007.44 ◽

2007 ◽

Cited By ~ 24

Author(s):

Eugene Zilberg ◽

Zheng Ming Xu ◽

David Burton ◽

Murrad Karrar ◽

Saroj Lal

Keyword(s):

Drowsiness Detection ◽

Detection Systems ◽

Non Invasive ◽

Driver Drowsiness ◽

Initial Analysis

Download Full-text

Frames of reference and their neural correlates within navigation in a 3D environment

Visual Neuroscience ◽

10.1017/s0952523812000168 ◽

2012 ◽

Vol 29 (3) ◽

pp. 183-191 ◽

Cited By ~ 1

Author(s):

MICHAL VAVREČKA ◽

VÁCLAV GERLA ◽

LENKA LHOTSKÁ ◽

MARTIN BRUNOVSKÝ

Keyword(s):

Three Dimensional ◽

Vertical Plane ◽

Neural Correlates ◽

Eeg Signal ◽

Lower Error Rate ◽

Signal Features ◽

Allocentric Reference Frame ◽

Frame Processing ◽

Electroencephalogram Eeg ◽

Eeg Features

AbstractThe goal of this study was an administration of the navigation task in a three-dimensional virtual environment to localize the electroencephalogram (EEG) features responsible for egocentric and allocentric reference frame processing in a horizontal and also in a vertical plane. We recorded the EEG signal of a traverse through a virtual tunnel to search for the best signal features that discriminate between specific strategies in particular plane. We identified intrahemispheric coherences in occipital–parietal and temporal–parietal areas as the most discriminative features. They have 10% lower error rate compared to single electrode features adopted in previous studies. The behavioral analysis revealed that 11% of participants switched from egocentric to allocentric strategy in a vertical plane, while 24% of participants consistently adopted egocentric strategy in both planes.

Download Full-text

Driver Drowsiness Detection System using Machine Learning

International Journal for Research in Applied Science and Engineering Technology ◽

10.22214/ijraset.2021.33888 ◽

2021 ◽

Vol 9 (VI) ◽

pp. 2270-2272

Author(s):

Charan M

Keyword(s):

Real Time ◽

Web Application ◽

Detection System ◽

Smart Phone ◽

Learning Approaches ◽

Drowsiness Detection ◽

Driver Drowsiness ◽

The Face ◽

Bounding Boxes ◽

Traditional Image

We propose a Driver drowsiness detection system, the purposes of which are to prevent from dangerous cause and to avoid accidents. Since all the processes on image recognition performed on a smart phone, the system does not need to send images to a server and runs on an android smart phone in a real-time way. Automatic image-based recognition is a particularly challenging task. Traditional image analysis approaches have achieved low classification accuracy in the past, whereas deep learning approaches without human supervision real-time drowsiness detection. This model classifies whether the person’s eyes are opened or closed. To recognize the face, a user should have to adjust camera such a way that it covers his face first, and then the system starts recognition within the indicated bounding boxes. In addition, the system estimates the actions of the person. This recognition process is performed repeatedly about every second. We will implement this system as Web application effectively for real-time recognition.

Download Full-text

Distinguishing Different Emotions Evoked by Music via Electroencephalographic Signals

Computational Intelligence and Neuroscience ◽

10.1155/2019/3191903 ◽

2019 ◽

Vol 2019 ◽

pp. 1-18 ◽

Cited By ~ 5

Author(s):

Yimin Hou ◽

Shuaiqi Chen

Keyword(s):

Recognition Accuracy ◽

Feature Selection Method ◽

Support Vector ◽

Eeg Signal ◽

Feature Sets ◽

Signal Features ◽

Correlation Based Feature Selection ◽

Series Of Experiments ◽

Electroencephalogram Eeg ◽

Eeg Features

Music can evoke a variety of emotions, which may be manifested by distinct signals on the electroencephalogram (EEG). Many previous studies have examined the associations between specific aspects of music, including the subjective emotions aroused, and EEG signal features. However, no study has comprehensively examined music-related EEG features and selected those with the strongest potential for discriminating emotions. So, this paper conducted a series of experiments to identify the most influential EEG features induced by music evoking different emotions (calm, joy, sad, and angry). We extracted 27-dimensional features from each of 12 electrode positions then used correlation-based feature selection method to identify the feature set most strongly related to the original features but with lowest redundancy. Several classifiers, including Support Vector Machine (SVM), C4.5, LDA, and BPNN, were then used to test the recognition accuracy of the original and selected feature sets. Finally, results are analyzed in detail and the relationships between selected feature set and human emotions are shown clearly. Through the classification results of 10 random examinations, it could be concluded that the selected feature sets of Pz are more effective than other features when using as the key feature set to classify human emotion statues.

Download Full-text