Command Recognition Based on Electroencephalography
碩士 === 國立臺北科技大學 === 電腦與通訊研究所 === 100 === This thesis investigates how to recognize a user''s intentions from a set of commands for machine control by measuring and analyzing his/her brainwaves. Our strategy is to convert the problem of an N-class decision into N binary decisions....
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2012
|
Online Access: | http://ndltd.ncl.edu.tw/handle/8n66g3 |
id |
ndltd-TW-100TIT05652071 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-100TIT056520712019-05-15T20:51:53Z http://ndltd.ncl.edu.tw/handle/8n66g3 Command Recognition Based on Electroencephalography 利用腦波訊號識別指令 Wen-Bin Jheng 鄭旺彬 碩士 國立臺北科技大學 電腦與通訊研究所 100 This thesis investigates how to recognize a user''s intentions from a set of commands for machine control by measuring and analyzing his/her brainwaves. Our strategy is to convert the problem of an N-class decision into N binary decisions. For example, in a task of identifying ten commands including digits from 0 to 9, our system prompts each digit for a user, and then analyzes his/her brainwave, thereby determining if the digit is the user''s intention. It is assumed that if the user''s intention is, say digit 7, then the resulting electroencephalogram (EEG) measured from the user should present a certain pattern of "Yes" when the system prompts digit 7, and present a certain pattern of "No" when the system prompts a digit other than 7. Therefore, the goal of our system is to determine if the user''s intention is "Yes" or not based on the measured EEG. Although there are several prior studies discussing such a binary EEG classification problem, all of them use 32 or more channels EEG to develop their systems, which involve inconvenient and uncomfortable recording head nets as well as expensive equipment, and therefore unsuitable for real applications of human-machine interface. In contrast, this study uses a simple, portable, and cheap instrument that extracts single-channel EEG from a user''s frontal lobe. The underlying beta waves of EEG are then distilled and used as the feature to determine a user''s intention. Our experiments conducted using 40 test EEG samples from 10 subjects show that the recognition accuracy obtained with our system is around 73%. Wei-Ho Tsai 蔡偉和 2012 學位論文 ; thesis 66 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺北科技大學 === 電腦與通訊研究所 === 100 === This thesis investigates how to recognize a user''s intentions from a set of commands for machine control by measuring and analyzing his/her brainwaves. Our strategy is to convert the problem of an N-class decision into N binary decisions. For example, in a task of identifying ten commands including digits from 0 to 9, our system prompts each digit for a user, and then analyzes his/her brainwave, thereby determining if the digit is the user''s intention. It is assumed that if the user''s intention is, say digit 7, then the resulting electroencephalogram (EEG) measured from the user should present a certain pattern of "Yes" when the system prompts digit 7, and present a certain pattern of "No" when the system prompts a digit other than 7. Therefore, the goal of our system is to determine if the user''s intention is "Yes" or not based on the measured EEG. Although there are several prior studies discussing such a binary EEG classification problem, all of them use 32 or more channels EEG to develop their systems, which involve inconvenient and uncomfortable recording head nets as well as expensive equipment, and therefore unsuitable for real applications of human-machine interface. In contrast, this study uses a simple, portable, and cheap instrument that extracts single-channel EEG from a user''s frontal lobe. The underlying beta waves of EEG are then distilled and used as the feature to determine a user''s intention. Our experiments conducted using 40 test EEG samples from 10 subjects show that the recognition accuracy obtained with our system is around 73%.
|
author2 |
Wei-Ho Tsai |
author_facet |
Wei-Ho Tsai Wen-Bin Jheng 鄭旺彬 |
author |
Wen-Bin Jheng 鄭旺彬 |
spellingShingle |
Wen-Bin Jheng 鄭旺彬 Command Recognition Based on Electroencephalography |
author_sort |
Wen-Bin Jheng |
title |
Command Recognition Based on Electroencephalography |
title_short |
Command Recognition Based on Electroencephalography |
title_full |
Command Recognition Based on Electroencephalography |
title_fullStr |
Command Recognition Based on Electroencephalography |
title_full_unstemmed |
Command Recognition Based on Electroencephalography |
title_sort |
command recognition based on electroencephalography |
publishDate |
2012 |
url |
http://ndltd.ncl.edu.tw/handle/8n66g3 |
work_keys_str_mv |
AT wenbinjheng commandrecognitionbasedonelectroencephalography AT zhèngwàngbīn commandrecognitionbasedonelectroencephalography AT wenbinjheng lìyòngnǎobōxùnhàoshíbiézhǐlìng AT zhèngwàngbīn lìyòngnǎobōxùnhàoshíbiézhǐlìng |
_version_ |
1719106175360827392 |