作者: Syed Khairul Bashar , Mohammed Imamul Hassan Bhuiyan
DOI: 10.1109/ICEEICT.2015.7307509
关键词:
摘要: In this paper, a method to classify arm movements using statistical features of electroencephalogram (EEG) signals calculated from wavelet packet and Fourier transforms, has been proposed. The EEG are analyzed bi-orthogonal family. transform is then applied the corresponding detail coefficients higher order moment named kurtosis magnitude components. shown be distinguishable for four different movements. K-nearest neighbor (KNN)-based classifiers developed these identify movements, right hand forward backward; left backward. A mean accuracy 92.84% achieved which better than some existing techniques.