脳波による掌握・発声動作想起を用いた言語選択の提案

カテゴリ: 部門大会

論文No: OS5-6

グループ名: 【C】2021年電気学会電子・情報・システム部門大会

発行日: 2021/09/08

タイトル(英語): A Proposal for Language Selection Using EEG-based Imaginations of Holding hands and Vocalization

著者名: 十亀 太雅（徳島大学）,伊藤 伸一（徳島大学）,伊藤 桃代（徳島大学）,福見 稔（徳島大学）

著者名(英語): Taiga Sogame (Tokushima University),Shinichi Ito (Tokushima University),Momoyo Ito (Tokushima University),Minoru Fukumi (Tokushima University)

キーワード: 脳波|事象関連脱同期|運動想起|閉じ込め症候群|k近傍法律動|electroencephalogram|event-related desynchronization|motor imagination|locked-in syndrome|k nearest-neighborrhythmic movement

要約(日本語): In this paper, we identify motor imagery such as hand-holding and vocalization. We propose a Japanese selection method using a table we made ourselves. The proposed method consists of EEG measurement, EEG feature extraction, and motor image classification. For the EEG measurement, we use a simple electroencephalograph. As EEG features, we use Fast Fourier Transform (FFT) to extract Event Related Synchronization (ERS) and Event Related Desynchronization (ERD) by alpha and beta waves(α and β waves). Support Vector Machine (SVM) is used to classify the motor images such as holding hands and vocalization. To show the effectiveness of the proposed method, we conducted experiments under real conditions and analyzed the results.

PDFファイルサイズ: 512 Kバイト