Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG
One way to recognize hand gestures is to use signal electromyography (EMG). The processed signal can use the time domain, frequency domain, or a mixture of the two domains. Meanwhile, the classification method that is widely used recently is the classification of Artificial Neural Networks (ANN). Th...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Universitas Syiah Kuala
2021-03-01
|
Series: | Jurnal Rekayasa Elektrika |
Subjects: | |
Online Access: | http://jurnal.unsyiah.ac.id/JRE/article/view/16844 |
id |
doaj-f5ddfc062c20481896ef5f65ecb37863 |
---|---|
record_format |
Article |
spelling |
doaj-f5ddfc062c20481896ef5f65ecb378632021-04-07T08:33:15ZengUniversitas Syiah KualaJurnal Rekayasa Elektrika1412-47852252-620X2021-03-01171364110.17529/jre.v17i1.1684412467Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMGDaniel Pamungkas0Sumantri R Kurniawan1Benrico F Simamora2Politeknik Negeri BatamPoliteknik Negeri BatamPoliteknik Negeri BatamOne way to recognize hand gestures is to use signal electromyography (EMG). The processed signal can use the time domain, frequency domain, or a mixture of the two domains. Meanwhile, the classification method that is widely used recently is the classification of Artificial Neural Networks (ANN). This paper presents a comparison study between time domains with frequency domain for EMG signals using ANN classification. This comparison aims to find out a better method for controlling the hand robot. The time domain features are root mean square (RMS) of the signal, while the signal’s octave band becomes a feature of the frequency domain. The EMG signals were obtained from the subject with eight fingers gestures. The results of this classification are used to control the robot’s hand. The success of each method in recognizing hand movements was counted. In addition, the response speed of the robot in changing positions is measured. The results showed that features using the frequency domain had a higher percentage of success than another domain. But the speed and memory used then the system using signals in the time domain is better.http://jurnal.unsyiah.ac.id/JRE/article/view/16844classificationelectromyographyfrequency domaintime domainneural network |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Daniel Pamungkas Sumantri R Kurniawan Benrico F Simamora |
spellingShingle |
Daniel Pamungkas Sumantri R Kurniawan Benrico F Simamora Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG Jurnal Rekayasa Elektrika classification electromyography frequency domain time domain neural network |
author_facet |
Daniel Pamungkas Sumantri R Kurniawan Benrico F Simamora |
author_sort |
Daniel Pamungkas |
title |
Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG |
title_short |
Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG |
title_full |
Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG |
title_fullStr |
Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG |
title_full_unstemmed |
Perbandingan Antara Domain Waktu dan Frekuensi untuk Pengenalan Sinyal EMG |
title_sort |
perbandingan antara domain waktu dan frekuensi untuk pengenalan sinyal emg |
publisher |
Universitas Syiah Kuala |
series |
Jurnal Rekayasa Elektrika |
issn |
1412-4785 2252-620X |
publishDate |
2021-03-01 |
description |
One way to recognize hand gestures is to use signal electromyography (EMG). The processed signal can use the time domain, frequency domain, or a mixture of the two domains. Meanwhile, the classification method that is widely used recently is the classification of Artificial Neural Networks (ANN). This paper presents a comparison study between time domains with frequency domain for EMG signals using ANN classification. This comparison aims to find out a better method for controlling the hand robot. The time domain features are root mean square (RMS) of the signal, while the signal’s octave band becomes a feature of the frequency domain. The EMG signals were obtained from the subject with eight fingers gestures. The results of this classification are used to control the robot’s hand. The success of each method in recognizing hand movements was counted. In addition, the response speed of the robot in changing positions is measured. The results showed that features using the frequency domain had a higher percentage of success than another domain. But the speed and memory used then the system using signals in the time domain is better. |
topic |
classification electromyography frequency domain time domain neural network |
url |
http://jurnal.unsyiah.ac.id/JRE/article/view/16844 |
work_keys_str_mv |
AT danielpamungkas perbandinganantaradomainwaktudanfrekuensiuntukpengenalansinyalemg AT sumantrirkurniawan perbandinganantaradomainwaktudanfrekuensiuntukpengenalansinyalemg AT benricofsimamora perbandinganantaradomainwaktudanfrekuensiuntukpengenalansinyalemg |
_version_ |
1721536250295353344 |