A dual-class voting mechanism for brain computer interface based on wavelet packet and support vector machine

2013 ◽  
Author(s):  
Guang Yang ◽  
Kenji Nakayama ◽  
Akihiro Hirano
Keyword(s):  
Support Vector Machine ◽  
Brain Computer Interface ◽  
Wavelet Packet ◽  
Computer Interface ◽  
Support Vector ◽  
Class Voting ◽  
Dual Class ◽  
Voting Mechanism

Practical Artifact Removal Brain-Computer Interface System

2014 ◽  
pp. 265-277
Author(s):  
Wei-Yen Hsu
Keyword(s):  
Support Vector Machine ◽  
Classification Accuracy ◽  
Brain Computer Interface ◽  
Computer Interface ◽  
Support Vector ◽  
Artifact Removal ◽  
Eeg Data ◽  
Interface System ◽  
Electroencephalogram Eeg ◽  
Bci System

In this chapter, a practical artifact removal Brain-Computer Interface (BCI) system for single-trial Electroencephalogram (EEG) data is proposed for applications in neuroprosthetics. Independent Component Analysis (ICA) combined with the use of a correlation coefficient is proposed to remove the EOG artifacts automatically, which can further improve classification accuracy. The features are then extracted from wavelet transform data by means of the proposed modified fractal dimension. Finally, Support Vector Machine (SVM) is used for the classification. When compared with the results obtained without using the EOG signal elimination, the proposed BCI system achieves promising results that will be effectively applied in neuroprosthetics.

EEG Feature Extraction and Pattern Classification Based on Motor Imagery in Brain-Computer Interface

2011 ◽  
Vol 3 (3) ◽  
pp. 43-56 ◽  
Author(s):  
Ling Zou ◽  
Xinguang Wang ◽  
Guodong Shi ◽  
Zhenghua Ma
Keyword(s):  
Support Vector Machine ◽  
Motor Imagery ◽  
Brain Computer Interface ◽  
Average Power ◽  
Computer Interface ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Classification Rate ◽  
Linear Discriminant ◽  
Reconstructed Signal

Accurate classification of EEG left and right hand motor imagery is an important issue in brain-computer interface. Firstly, discrete wavelet transform method was used to decompose the average power of C3 electrode and C4 electrode in left-right hands imagery movement during some periods of time. The reconstructed signal of approximation coefficient A6 on the sixth level was selected to build up a feature signal. Secondly, the performances by Fisher Linear Discriminant Analysis with two different threshold calculation ways and Support Vector Machine methods were compared. The final classification results showed that false classification rate by Support Vector Machine was lower and gained an ideal classification results.

scholarly journals Single channel electroencephalogram (EEG) brain computer interface (BCI) feature extraction and quantization method for support vector machine classification

2018 ◽  
Vol 7 (4) ◽  
pp. 2095 ◽  
Author(s):  
Tarmizi Ahmad Izzuddin ◽  
Norlaili Mat Safri ◽  
Fauzal Naim Zohedi ◽  
Mohamad Afzan Othman ◽  
Muhammad Shaufil Adha Shawkany Hazim
Keyword(s):  
Support Vector Machine ◽  
Single Channel ◽  
Brain Computer Interface ◽  
Target Object ◽  
International Standards ◽  
Assistive Robotics ◽  
Computer Interface ◽  
Support Vector ◽  
Electroencephalogram Eeg ◽  
Bci System

Over the recent years, there has been a huge interest towards Electroencephalogram (EEG) based brain computer interface (BCI) system. BCI system enables the extraction of meaningful information directly from the human brain via suitable signal processing and machine learning method and thus, many researches have applied this technology towards rehabilitation and assistive robotics. Such application is important towards improving the lives of people with motor diseases such as Amytrophic Lateral Scelorosis (ALS) disease or people with quadriplegia/tetraplegia. This paper introduces features extraction method based on the Fast Fourier Transform (FFT) with logarithmic bin-ning for rapid classification using Support Vector Machine (SVM) algorithm, with an application towards a BCI system with a shared con-trol scheme. In general, subjects wearing a single channel EEG electrode located at F8 (10-20 international standards) were required to syn-chronously imagine a star rotating and mind relaxation at specific time and direction. The imagination of a star would trigger a mobile robot suggesting that there exists a target object at certain direction. Based on the proposed algorithm, we showed that our algorithm can distin-guish between mind relaxation and mental star rotation with up to 80% accuracy from the single channel EEG signals.  

scholarly journals Improved Transductive Support Vector Machine for a Small Labelled Set in Motor Imagery-Based Brain-Computer Interface

2019 ◽  
Vol 2019 ◽  
pp. 1-16 ◽  
Author(s):  
Yilu Xu ◽  
Jing Hua ◽  
Hua Zhang ◽  
Ronghua Hu ◽  
Xin Huang ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Motor Imagery ◽  
Optimization Problem ◽  
Brain Computer Interface ◽  
The Other ◽  
Computer Interface ◽  
Geometric Feature ◽  
Support Vector ◽  
Transductive Support Vector Machine ◽  
Calibration Time

Long and tedious calibration time hinders the development of motor imagery- (MI-) based brain-computer interface (BCI). To tackle this problem, we use a limited labelled set and a relatively large unlabelled set from the same subject for training based on the transductive support vector machine (TSVM) framework. We first introduce an improved TSVM (ITSVM) method, in which a comprehensive feature of each sample consists of its common spatial patterns (CSP) feature and its geometric feature. Moreover, we use the concave-convex procedure (CCCP) to solve the optimization problem of TSVM under a new balancing constraint that can address the unknown distribution of the unlabelled set by considering various possible distributions. In addition, we propose an improved self-training TSVM (IST-TSVM) method that can iteratively perform CSP feature extraction and ITSVM classification using an expanded labelled set. Extensive experimental results on dataset IV-a from BCI competition III and dataset II-a from BCI competition IV show that our algorithms outperform the other competing algorithms, where the sizes and distributions of the labelled sets are variable. In particular, IST-TSVM provides average accuracies of 63.25% and 69.43% with the abovementioned two datasets, respectively, where only four positive labelled samples and sixteen negative labelled samples are used. Therefore, our algorithms can provide an alternative way to reduce the calibration time.

Sign in / Sign up

Export Citation Format

Share Document