A Brain Computer Interface Based on FFT and Multilayer Neural Network : Feature Extraction and Generalization

A Brain Computer Interface Based on FFT and Multilayer Neural Network : Feature Extraction and Generalization

フォーマット:

論文

責任表示:

Kaneda, Yasuaki ; Nakayama, Kenji ; Hirano, Akihiro

言語:

英語

出版情報:

IEICE The Institute of Electronics, Information and Communication Engineers, 2007-01-01

著者名:

掲載情報:

電子情報通信学会技術研究報告. VLD, VLSI設計技術 = Technical report of IEICE. VLD

ISSN:

0913-5685

巻:

107

通号:

103

開始ページ:

1

終了ページ:

6

バージョン:

publisher

概要:

金沢大学理工研究域　電子情報学系<br />脳波のFFTと階層形ニューラルネットワークを用いるブレイン・コンピュータ・インタフェイス(BCI)に関して,以前に前処理の方法をいくつか提案し,メンタルタスクの分類性能を向上した.本稿では,まず,階層形ニューラルネットワークでメンタルタスクを分類するために用いられる特徴の解析を行った.特徴は結合荷重の分布に基づいて解析した.隠れ層から出力層への結合荷重はメンタルタスクに対して独立になる傾向があった.従って,入力層から各メ … ンタルタスクに対応する隠れユニットへの結合荷重分布がメンタルタスク毎の特徴を表している.次に,汎化能力を向上する2通りの学習法について検討を行った.一つは,ニューラルネットワークの入力データに乱数を加える方法であり,もう一つは,結合荷重を圧縮する方法する方法である.シミュレーションの結果,いずれの方法もテストデータに対する分類性能を向上することが出来たが,乱数を加える方法が有効であることが分かった. In this paper, a multilayer neural network is applied to 'Brain Computer Interface' (BCI), which is one of hopeful interface technologies between humans and machines. Amplitude of the FFT of the brain waves are used for the input data. Several techniques have been introduced for pre-processing the brain waves. They include segmentation along the time axis for fast response, nonlinear normalization to emphasize important information, averaging samples of the brain waves to suppress noise effects, reduction in the number of the samples to realize a small size network, and so on. In this paper, two kinds of generalization techniques, including adding small random noises to the input data and decaying connection weight magnitude, are applied. Their usefulness are analyzed and compared base on correct and error classifications. Simulation is carried out by using the brain waves, which are available from the web site of Colorado State University. The number of mental tasks is five. Some data sets are used for training the multilayer neural network, and the remaining data sets are used for testing. In our previous work, classification accuracy of 64%〜74% for the test data have been achieved. In this paper, by applying the generalization techniques, the accuracy can be improved up to 80%～88%. 続きを見る

URL:

http://hdl.handle.net/2297/18406

タイトル・著者・出版者が同じ資料

詳細

主題:	Brain computer interface; Feature extraction; Generalization; Mental task; Multilayer neural network; ブレイン・コンピュータ・インタフェイス; メンタルタスク; 汎化能力; 特殊抽出; 階層形ニューラルネットワーク
受理日:	2007-01-01

類似資料:

1

2

3

4

5

6

7

8

9

10

11

12

1 論文 A brain computer interface based on FFT and multilayer neural network -feature extraction and generalization- Nakayama, Kenji, Kaneda, Yasuaki, Hirano, Akihiro IEEE = Institute of Electrical and Electronics Engineers	7 論文 A brain computer interface based on neural network with efficient pre-processing Nakayama, Kenji, Inagaki, Kiyoto IEEE = Institute of Electrical and Electronics Engineers
2 論文 Neural network based BCI by using orthogonal components of multi-channel brain waves and generalization Nakayama, Kenji, Horita, Hiroki, Hirano, Akihiro Springer Verlag	8 論文 A brain computer interface based on neural network with efficient pre-processing Nakayama, Kenji, Inagaki, Kiyoto IEEE = Institute of Electrical and Electronics Engineers
3 論文 A bci using megvision and multilayer neural network - Channel optimization and main lobe contribution analysis Nakayama, Kenji, Kaneda, Yasuaki, Hirano, Akihiro, Haruta, Yasuhiro IEEE = Institute of Electrical and Electronics Engineers	9 論文 A selective learning algorithm for nonlinear synapses in multilayer neural networks Nakayama, Kenji, Hirano, Akihiro, Fusakawa, M. IEEE(Institute of Electrical and Electronics Engineers)
4 論文 Optimization of activation functions in multilayer neural network applied to pattern classification Nakayama, Kenji, Kimura, Yoshinori IEEE(Institute of Electrical and Electronics Engineers)	10 論文 A Multilayer Neural Network with Nonlinear Inputs and Trainable Activation Functions: Structure and Simultaneous Learning Algorithm Nakayama, Kenji, Hirano, Akihiro, Ido, Issei IEEE(Institute of Electrical and Electronics Engineers)
5 論文 On generalization of multilayer neural network applied to predicting protein secondary structure Nakayama, Kenji, Hirano, Akihiro, Fukumura, K. IEEE(Institute of Electrical and Electronics Engineers)	11 論文 Training data selection method for generalization by multilayer neural networks Hara, Kazuyuki, Nakayama, Kenji
6 論文 A brain computer interface based on neural network with efficient pre-processing Nakayama, Kenji, Inagaki, Kiyoto IEEE(Institute of Electrical and Electronics Engineers)	12 論文 Selection of minimum training sata for generalization and on-line training by multilayer neural networks Hara, Kazuyuki, Nakayama, Kenji IEEE(Institute of Electrical and Electronics Engineers)