EEG Based Classification of Hand Movements using BCI

Abstract
Authors
Keywords
Conclusion
References

Brain interface computer (BCI) is new area of disabled people. Detection of imagination of left is made in my project and it can used for external devices such as robotic arm. The electrical activity can be picked up from scalp electroencephalogram electrodes. Here we have collected signals using Enobio 8 software. The collected signals are given to NE_Viewer filter to get Filtered data. The filtered data is then taken to extract from time domain information in both ALPHA and BETA bands using wavelet transform. The ALPHA and BETA waves are used and extracted some of the parameters such as power, standard deviation, and variance. The LDA classifier is used to classify the Imagination of left and right hand movements. In our project we have taken 40 samples for training data set and for testing phase we have taken 10 samples for testing and we have achieved 90% of accuracy.

Published In : IJCSN Journal Volume 5, Issue 4

Date of Publication : August 2016

Pages : 687-691

Figures :03

Tables : 01

Lavanya T H : PG Student, Department of CS&E, Channabasaveshwara Institute of Technology, TUMKUR India.

Jyothi K S : Associate Professor Department of CS&E, Channabasaveshwara Institute of technology, TUMKUR, India.

EEG, BCI

In our project we are only concentrating on classification of hand movements of human body, left and right hand movement. We instruct the subjects in the beginning only to focus on these hand movements. The signals were recorded from eight channels according the 10-20 international system. We are using NIC software for collecting and storing EEG samples. The collected signals were pre-processed and the artifacts were removed by designing Parks-McClellan FIR filter. We have recorded samples from participants and divided those samples into two sets, the training set and the testing set. The training set consists of samples of 29 participants and testing set consists of 5 participants. We use wavelet transform method for extracting the features, power, standard deviation, variance, these features are used in classification method. Here we are using Linear Discreminant Analysis (LDA) to classify hand movements.

[1] Automated Classification of L/R Hand Movement EEG Signals using Advanced Feature Extraction and Machine Learning-Mohammad H. Alomari, AyaSamaha, [2] Classification of EEG Signals Recorded During Right/Left Hand Movement Imagery Using Fourier Transform Based Features-A.B.M. AowladHossain, Md. WasiurRahman, ManjurulAhsanRiheen. [3] Classification of EEG Signal for Left and Right Wrist Movements Using AR Modelling- OnderAydemir, TemelKayikçioglu [4] Classification of Left/Right Hand Movement from EEG Signal by Intelligent Algorithms-Mohammed Hassan,Mohamed I. Eladawy, Ahmed FaragSeddik [5] Classification of EEG Signals Based on Imaginary Movement of Right and Left Hand Wrist-Saugat Bhattacharyya, AnweshaKhasnobish, AmitKonar, D.N Tibarewala1, Atulya K. Nagar.