文章基本信息

标题：EEG BASED USER IDENTIFICATION METHODS USING TWO SEPARATE SETS OF FEATURES BASED ON DCT AND WAVELET
本地全文：下载
作者：HEND A. HADI ; DR. LOAY E. GEORGE
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：22
页码：6116
出版社：Journal of Theoretical and Applied
摘要：Brain activities may represented by EEG signals, which are set of measures using electrodes along the scalp, they are more secret, sensitive, and hard to steal and reproduce. They hold great potential to provide robust and secure biometric system for user identification and verification. This paper aims to present a comparison between our previous proposed feature set that based on Partitioned Fourier spectra and some new features sets proposed in this work. They established as simple, fast, and promising set of features for EEG-based identification system. The first introduced feature set is based on the energy distribution of DCT AC-components, and the second set is the statistical moments for three types of wavelet transforms. Each set of features is tested using normalized distance measures for matching stage. Each proposed method was tested using the publicly available EEG CSU dataset which was collected from seven healthy volunteers and the publicly available EEG Motor Movement/Imagery dataset which is relatively large dataset was collected from 109 healthy subjects. The attained identification results are encouraging with best recognition rate is (100%) for all proposed methods and for both datasets. All tested feature sets were extracted under the condition, which was adopted in our previous work, that is "they should extracted from EEG data belong to single task & signal channel". All achieved results are considered competitive when compared with the results of other recently published works. The adopted condition reduced the computational complexity and thus reduced the required processing time. Also, the wavelets based methods hold computational complexity less than DFT and DCT with recognition rates are more competitive to them.
关键词：Associate Professor. Faculty of Electrical Engineering; Universiti Teknikal Malaysia Melaka (UTeM);Hang Tuah Jaya; 76100 Durian Tunggal; Melaka; Malaysia