296
Asian Journal of Engineering, Sciences and Technology - Volume 5, Issue 2 2015
By Jamal Ahmed, Abdul Majid, Idress Riaz, Sarah Bano , M. Anas & M.Ghazaal
Keywords: BCI, Electencephlography, Motor Neurons Disease, Artificial Neural Network, Multi-Layer Perceptron. IEMG, MAV, SVM
Non-invasive Electroencephalography (EEG) based Brain Computer Interfaces (BCIs) is a growing technology which provides a possibility for intuitive operation comprising a multi-degree of freedom for upper extremity prosthesis. In this paper, we investigate bio signal classification concerns. Understanding bio signals and perform any action on behalf of it is a difficult task to accomplish. This method comprises of classification, feature extraction of EEG signals and objective was to make process as simple as possible. In this trial we first collected the patterns of four different types of facial activities namely right smirk, left smirk, eye blink and neutral activity. Firstly Samples were preprocessed. Wavelet decomposition is used to extract features from signal and IEMG and MAV are used to make feature vectors. Multilayer perceptron (MLP) and Support vector machines are used for classification. Results suggest that MLP can perform classification task at high accuracy. Our proposed model facilitates patient who are suffering from severe muscle injury and disability. They can freely move around by means of facial muscles regardless of any assistance.