scholarly journals EEG-Based 3D Visual Fatigue Evaluation Using CNN

Electronics ◽  
2019 ◽  
Vol 8 (11) ◽  
pp. 1208 ◽  
Author(s):  
Kang Yue ◽  
Danli Wang
Keyword(s):  
Feature Extraction ◽  
Temporal Structure ◽  
Visual Fatigue ◽  
Eeg Signals ◽  
Visual Evaluation ◽  
Time Frequency ◽  
Multi Scale ◽  
Temporal Features ◽  
Depth Analysis ◽  
Fatigue Evaluation

Visual fatigue evaluation plays an important role in applications such as virtual reality since the visual fatigue symptoms always affect the user experience seriously. Existing visual evaluation methods require hand-crafted features for classification, and conduct feature extraction and classification in a separated manner. In this paper, we conduct a designed experiment to collect electroencephalogram (EEG) signals of various visual fatigue levels, and present a multi-scale convolutional neural network (CNN) architecture named MorletInceptionNet to detect visual fatigue using EEG as input, which exploits the spatial-temporal structure of multichannel EEG signals. Our MorletInceptionNet adopts a joint space-time-frequency features extraction scheme in which Morlet wavelet-like kernels are used for time-frequency raw feature extraction and inception architecture are further used to extract multi-scale temporal features. Then, the multi-scale temporal features are concatenated and fed to the fully connected layer for visual fatigue evaluation using classification. In experiment evaluation, we compare our method with five state-of-the-art methods, and the results demonstrate that our model achieve overally the best performance better performance for two widely used evaluation metrics, i.e., classification accuracy and kappa value. Furthermore, we use input-perturbation network-prediction correlation maps to conduct in-depth analysis into the reason why the proposed method outperforms other methods. The results suggest that our model is sensitive to the perturbation of β (14–30 Hz) and γ (30–40 Hz) bands. Furthermore, their spatial patterns are of high correlation with that of the corresponding power spectral densities which are used as evaluation features traditionally. This finding provides evidence of the hypothesis that the proposed model can learn the joint time-frequency-space features to distinguish fatigue levels automatically.

scholarly journals Motor Imagery EEG Signal Recognition Using Deep Convolution Neural Network

2021 ◽  
Vol 15 ◽  
Author(s):  
Xiongliang Xiao ◽  
Yuee Fang
Keyword(s):  
Feature Extraction ◽  
Motor Imagery ◽  
Signal To Noise Ratio ◽  
Large Individual ◽  
Eeg Signal ◽  
Signal Recognition ◽  
Eeg Signals ◽  
Time Frequency ◽  
Morlet Wavelet Transform

Brain computer interaction (BCI) based on EEG can help patients with limb dyskinesia to carry out daily life and rehabilitation training. However, due to the low signal-to-noise ratio and large individual differences, EEG feature extraction and classification have the problems of low accuracy and efficiency. To solve this problem, this paper proposes a recognition method of motor imagery EEG signal based on deep convolution network. This method firstly aims at the problem of low quality of EEG signal characteristic data, and uses short-time Fourier transform (STFT) and continuous Morlet wavelet transform (CMWT) to preprocess the collected experimental data sets based on time series characteristics. So as to obtain EEG signals that are distinct and have time-frequency characteristics. And based on the improved CNN network model to efficiently recognize EEG signals, to achieve high-quality EEG feature extraction and classification. Further improve the quality of EEG signal feature acquisition, and ensure the high accuracy and precision of EEG signal recognition. Finally, the proposed method is validated based on the BCI competiton dataset and laboratory measured data. Experimental results show that the accuracy of this method for EEG signal recognition is 0.9324, the precision is 0.9653, and the AUC is 0.9464. It shows good practicality and applicability.

scholarly journals Coherence analysis of EEG in locomotion using graphs

2017 ◽  
Author(s):  
◽  
G. Quiroz
Keyword(s):  
Assistive Devices ◽  
Brain Machine Interface ◽  
Coherence Analysis ◽  
Connectivity Analysis ◽  
Eeg Signals ◽  
Time Frequency ◽  
Temporal Features ◽  
Machine Interface ◽  
Spatio Temporal ◽  
Motion Intention

One of the most interesting brain machine interface (BMI) applications, is the control of assistive devices for rehabilitation of neuromotor pathologies. This means that assistive devices (prostheses, orthoses, or exoskeletons) are able to detect user motion intention, by the acquisition and interpretation of electroencephalographic (EEG) signals. Such interpretation is based on the time, frequency or space features of the EEG signals. For this reason, in this paper a coherence-based EEG study is proposed during locomotion that along with the graph theory allows to establish spatio-temporal parameters that are characteristic in this study. The results show that along with the temporal features of the signal it is possible to find spatial patterns in order to classify motion tasks of interest. In this manner, the connectivity analysis alongside graphs provides reliable information about the spatio-temporal characteristics of the neural activity, showing a dynamic pattern in the connectivity during locomotions tasks.

Visualization of EEG Using Time-Frequency Distributions

1997 ◽  
Vol 36 (04/05) ◽  
pp. 298-301 ◽  
Author(s):  
B. Stiber ◽  
S. Sato
Keyword(s):  
Brain Activity ◽  
Temporal Structure ◽  
Frequency Resolution ◽  
Signal Frequency ◽  
Time Varying ◽  
Eeg Analysis ◽  
Frequency Distributions ◽  
Time Frequency ◽  
Temporal Features ◽  
Time Varying Frequency

Abstract:The EEG is a time-varying or nonstationary signal. Frequency and amplitude are two of its significant characteristics, and are valuable clues to different states of brain activity. Detection of these temporal features is important in understanding EEGs. Commonly, spectrograms and AR models are used for EEG analysis. However, their accuracy is limited by their inherent assumption of stationarity and their trade-off between time and frequency resolution. We investigate EEG signal processing using existing compound kernel time-frequency distributions (TFDs). By providing a joint distribution of signal intensity at any frequency along time, TFDs preserve details of the temporal structure of the EEG waveform, and can extract its time-varying frequency and amplitude features. We expect that this will have significant implications for EEG analysis and medical diagnosis.

scholarly journals Multi-Scale 3D Convolution Network for Video Based Person Re-Identification

2019 ◽  
Vol 33 ◽  
pp. 8618-8625 ◽  
Author(s):  
Jianing Li ◽  
Shiliang Zhang ◽  
Tiejun Huang
Keyword(s):  
Feature Extraction ◽  
Feature Learning ◽  
The Other ◽  
Temporal Attention ◽  
Spatial Feature ◽  
Multi Scale ◽  
Temporal Cues ◽  
Temporal Features ◽  
State Of Art ◽  
Temporal Feature

This paper proposes a two-stream convolution network to extract spatial and temporal cues for video based person ReIdentification (ReID). A temporal stream in this network is constructed by inserting several Multi-scale 3D (M3D) convolution layers into a 2D CNN network. The resulting M3D convolution network introduces a fraction of parameters into the 2D CNN, but gains the ability of multi-scale temporal feature learning. With this compact architecture, M3D convolution network is also more efficient and easier to optimize than existing 3D convolution networks. The temporal stream further involves Residual Attention Layers (RAL) to refine the temporal features. By jointly learning spatial-temporal attention masks in a residual manner, RAL identifies the discriminative spatial regions and temporal cues. The other stream in our network is implemented with a 2D CNN for spatial feature extraction. The spatial and temporal features from two streams are finally fused for the video based person ReID. Evaluations on three widely used benchmarks datasets, i.e.,MARS, PRID2011, and iLIDS-VID demonstrate the substantial advantages of our method over existing 3D convolution networks and state-of-art methods.

scholarly journals A Comparative Analysis of Time-frequency Feature Extraction Techniques for Large Scale Electroencephalogram Data

2021 ◽  
Vol 10 (1) ◽  
pp. 14-24
Keyword(s):  
Feature Extraction ◽  
Large Scale ◽  
Extraction Methods ◽  
Kernel Functions ◽  
The Body ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Eeg Signals ◽  
Time Frequency ◽  
The Impact

Recognition of human emotions is a fascinating research field that motivates many researchers to use various approaches, such as facial expression, speech or gesture of the body. Electroencephalogram (EEG) is another approach of recognizing human emotion through brain signals and has offered promising findings. Although EEG signals provide detail information on human emotional states, the analysis of non-linear and chaotic characteristics of EEG signals is a substantial problem. The main challenge remains in analyzing EEG signals to extract relevant features in order to achieve optimum classification performance. Various feature extraction methods have been developed by researchers, which mainly can be categorized under time, frequency or time-frequency based feature extraction methods. Yet, there are numerous setting that could affect the performance of any model. In this paper, we investigated the performance of Discrete Wavelet Transform (DWT) and Discrete Wavelet Packet Transform (DWPT), which are time-frequency domain methods using Support Vector Machine (SVM) and k-Nearest Neighbor (KNN) classification techniques. Different SVM kernel functions and distance metrics of KNN are tested in this study by using subject-dependent and subject -independent approaches. The experiment is implemented using publicly available DEAP dataset. The experimental results show that DWT is mostly suitable with weighted KNN classifier while DWPT reported better results when tested using Linear SVM classifier to accurately classify the EEG signals on subject-dependent approach. Consistent results are observed for DWT-KNN on subject-independent approach, however SVM works better in the setting of quadratic kernel functions. These results indicate that further investigation is significant to examine the impact of different setting of methods in analyzing large scale of EEG data

scholarly journals Hybrid Feature Extraction Technique for Multi-Classification of Ictal and Non-Ictal EEG Epilepsy Signals

2020 ◽  
Vol 19 (2) ◽  
pp. 1-11
Author(s):  
Sani Saminu ◽  
Guizhi Xu ◽  
Shuai Zhang ◽  
Abd El Kader Isselmou ◽  
Adamu Halilu Jabire ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Extraction Technique ◽  
Support Vector ◽  
Single Channels ◽  
Eeg Signals ◽  
Feed Forward Neural Network ◽  
Hybrid Features ◽  
Time Frequency ◽  
Sensitivity Specificity ◽  
Feature Extraction Technique

These Electroencephalography (EEG) signals is an effective tool for identification, monitoring, and treatment of epilepsy, but EEG signals need highly experienced personnel to interpret it correctly due to its complexity, even for an expert it is monotonous and usually consume much time. Therefore, the automatic computer-aided device (CAD) needs to be developed to overcome those challenges associated with epilepsy interpretation and diagnosis. The system efficiency relies largely on the quality of features supply as input to classifiers. This paper presents an efficient feature extraction technique to develop a CAD system that can detect and classify normal, interictal and ictal epilepsy signals correctly with high accuracy. Our approach employs time-frequency features, statistical features and nonlinear features combined as hybrid features to train and test the classifier. Machine learning classifiers of multi-class support vector machine (mSVM) and feed-forward neural network (FFNN) with fivefold cross-validation are used to classifies normal, interictal and ictal with our proposed features. Our system was tested using a publicly available database with three classes each of 100 single channels EEG signals of 4096 samples point each. Based on sensitivity, specificity, and accuracy, our proposed approach of multiclass classification shows a good performance with 96.7%, 98.3% and 100% of sensitivity, specificity, and accuracy respectively.

Sign in / Sign up

Export Citation Format

Share Document