Mental and Motor Task Classification by LDA

Abstract

Electroencephalogram (EEG) is the easiest and the painless method to reveal the electrical activity of the brain tissue in order to understand the functioning of the brain and also for clinical diagnostics. When the mental differences identified from EEG signals, this could led to handicapped people to communicate with their surroundings. The work presented here is aimed to classify two different mental tasks and motor behaviors. The features are extracted by power spectral density method and a further step developed to choose six different features from power spectral densities. The generated feature vectors are transferred to the classifier. The classification is done with the classical Linear Discriminant Analysis method. A considerable difference value has been reached for mental tasks and the hemispheric changes. Due to the frequency changes at each electrode locations, a control methodology for the people who are lack of movement control can be developed.