| 大会名称 |
| 2016年 総合大会 |
| 大会コ－ド |
| 2016G |
| 開催年 |
|
2016
|
| 発行日 |
|
2016/3/1
|
| セッション番号 |
|
A-18
|
| セッション名 |
|
バイオメトリクス
|
| 講演日 |
|
2016/3/15
|
| 講演場所(会議室等) |
|
センター2号館 4F 2409
|
| 講演番号 |
|
A-18-3
|
| タイトル |
|
Experimental Results on EEG-based Person Identification with Machine Learning
|
| 著者名 |
|
○Felix Putra Sjamsudin, Mutsumi Suganuma, Wataru Kameyama,
|
| キーワード |
|
Authentication, EEG, Machine Learning, SVM
|
| 抄録 |
|
In our research, we investigate the use of Electroencephalogram (EEG) to identify different individuals. We perform SVM machine learning classification of different subject’s EEG based on 3 EEG features (Discrete Fourier Transform, Zero Crossing Rate, and Hjorth parameters). We use a 4-electrode EEG capturing device to capture the EEG of 15 subjects while performing two different tasks (relax and listening to music). The classification accuracy between each EEG feature and the combination of 3 features are compared. With all 3 features, we achieve 95.06% classification accuracy of subjects.
|
| 本文pdf |
|
PDF download
|
