مسار بولونيا
AR

محمد زكي حسين جاسم علي الفائز

السنة عنوان البحث نشر البحث
2018 ANALYSIS AND IMPLEMENTATION OF BRAINWAVES FEATURE EXTRACTION AND CLASSIFICATION TO CONTROL ROBOTIC HAND Iraqi Journal of Information and Communications Technology(IJICT) Vol. 1, Issue 3, November 2018
Abstract– In this paper, feature extraction methods such as time domain, frequency domain and spatial domain ere investigated. Where Mean Absolute Value (MAV), Integrated Absolute Value (IAV), Zero Crossing (ZC), Root ean Square (RMS), Waveform Length (WL) and Slope Sign Change (SSC) are the used time domain features. Autoregressive Feature (AR) is the frequency domain feature and the spatial domain feature is the Common patial Patterns (CSP). Channel selection algorithm was proposed for dimensionality reduction using Matlab code. esults of the above algorithm were compared with Matlab library of Principle Component Analysis (PCA). The xtracted feature vectors were fed into Support Vector Machine with Radial Basis Function kernel (SVM-RBF) to rain the classifier. The pair of algorithms (feature extraction plus dimensionality reduction) that owned the lowest lassification error rate were used to control a Humanoid Robotic Hand (HRH) in offline mode. EEG dataset of two lasses and three bipolar channels was used. Results showed that CSP features achieved the lowest classification rror rate for both dimensionality reduction techniques with 2.14%. Results recommends to use (CSP plus channel election algorithm) over (PCA plus PCA) since the former owned lowest classification processing time of 8.2s over .5s for the later.