WSEAS Transactions on Signal Processing
Print ISSN: 1790-5052, E-ISSN: 2224-3488
Volume 15, 2019
Human Identification based on electroencephalography Signals using Sample Entropy and Horizontal Visibility Graphs
Authors: , ,
Abstract: Biometric development depends on electroencephalography (EEG) distinguishes people by utilizing individual qualities in human brainwaves. Tow Essential features of EEG signals are Liveliness strength against adulteration. However , far reaching study on human authentication utilizing EEG signals is still remain. In this paper we propose a two-phase approach to distinguish EEG signals. The first phase , feature vectors are based on Sample Entropy (SaE) and Horizontal Visibility Graphs (HVG) to extract feature vector of EEG activities.The second phase performs a classification of these feature vectors using K-Nearest Neighbour (KNN) classifiers. We test the accuracy of the proposed approach on Machine Learning Repository (UCI) dataset . Experimental results on this dataset demonstrate significant improvement in the classification accuracy compared to other reported results. Our study applied two models, the first model using 13 channels to feature extraction . It was found that classifier with HVG had a much better performance giving the highest accuracy gave 94.8 % compared to classifier with SaE gave 83.7% accuracy . The second model using all channels. The classification accuracy with HVG gave 97.4% and with SaE gave 92.6% .
Search Articles
Keywords: Brain Computer Interface, Horizontal Visibility Graphs, Sample Entropy, Graph Entropy, EEG Signals, K-Nearest Neighbor, Machine Learning, Biometrics
Pages: 47-54
WSEAS Transactions on Signal Processing, ISSN / E-ISSN: 1790-5052 / 2224-3488, Volume 15, 2019, Art. #7