Journal ID : TRKU-07-03-2020-9694
[This article belongs to Volume - 62, Issue - 02]
Total View : 141

Title : Classification of EEG signal of Motor Imagery using Different Methods

Abstract :

The procedure of electroencephalography (EEG) is a famous process in neuroscience that is utilized for extracting the activity of brain signals related to voluntary and involuntary actions. The systems of BCI are capable of making the persons control any external device remotely utilizing the signals of the brain with no neurophysical interference. Many techniques used for isolating noise from EEG signals like filters and blind source separation. The main problem is Artifacts (noise) such as Electrocardiography (ECG), electromyography (EMG), electrooculography (EOG) and power line (LN). Added to   EEG signals and affect the performance of the system. Separating these signals into individual components to delete the artifacts is complicated. In this paper on the combination between stone-blind source separation and bandpass filter has been implemented bandpass filter was used and an algorithm based on a stone-blind source separation algorithm is presented for extracting the features from EEG based recorded brain signals. These features are extracted for Motor imagery; the function is the index finger motions to left or right. The proposed system also verified based on two Classifiers. The obtained results the precision rate were (77% by Hoeffding Tree Classifier) while the precision rate was (58% by Random forest)

Full article