An efficient approach to recognize hand gestures using machine-learning algorithms

Md Ferdous Wahid, Reza Tafreshi, Mubarak Al-Sowaidi, Reza Langari

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Electromyography (EMG) from a subject's upper limb can be used to train a machine-learning algorithm to classify different hand gestures. However, variability in the EMG signal due to between-subject differences can substantially degrade the machine-learning performance. This variation is usually due to the differences in both anatomical and physiological properties of the muscles, levels of muscle contraction, and inherent noises from the sensors. The aim of this study is to develop a subject-independent algorithm that can accurately classify different hand gestures. To minimize the between-subject differences, some selected time-domain EMG features are normalized to the area under the averaged root mean square curve (AUC-RMS). Five adult subjects with ages ranging 20-37 years performed three hand gestures including fist, wave-in, and wave-out for ten to twelve times each. Five machine-learning algorithms, including k-nearest neighbor (KNN), discriminant analysis (DA), Naïve Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM) were used to classify the three different hand gestures. The results showed that the EMG features were moderately to strongly correlated with the AUC-RMS values. The SVM yielded maximum classification accuracy using the original EMG features (97.56%) which was significantly improved by using the normalized EMG features (98.73%) (p<0.05). The accuracy distribution of all classifiers were found to be closer to mean values when using the normalized EMG features compared to using the original EMG features. The developed approach of classifying different hand gestures will be useful in biomedical applications such as controlling exoskeletons and in certain human-computer interaction settings.

Original languageEnglish
Title of host publication4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018
PublisherIEEE Computer Society
Pages171-176
Number of pages6
Volume2018-March
ISBN (Electronic)9781538614617
DOIs
Publication statusPublished - 3 Jul 2018
Event4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018 - Tunis, Tunisia
Duration: 28 Mar 201830 Mar 2018

Other

Other4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018
CountryTunisia
CityTunis
Period28/3/1830/3/18

Fingerprint

Electromyography
Learning algorithms
Learning systems
Support vector machines
Muscle
Discriminant analysis
Human computer interaction
Classifiers
Sensors

Keywords

  • electromyography
  • hand gestures
  • machine learning
  • pattern recognition

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Wahid, M. F., Tafreshi, R., Al-Sowaidi, M., & Langari, R. (2018). An efficient approach to recognize hand gestures using machine-learning algorithms. In 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018 (Vol. 2018-March, pp. 171-176). IEEE Computer Society. https://doi.org/10.1109/MECBME.2018.8402428

An efficient approach to recognize hand gestures using machine-learning algorithms. / Wahid, Md Ferdous; Tafreshi, Reza; Al-Sowaidi, Mubarak; Langari, Reza.

4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018. Vol. 2018-March IEEE Computer Society, 2018. p. 171-176.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wahid, MF, Tafreshi, R, Al-Sowaidi, M & Langari, R 2018, An efficient approach to recognize hand gestures using machine-learning algorithms. in 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018. vol. 2018-March, IEEE Computer Society, pp. 171-176, 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018, Tunis, Tunisia, 28/3/18. https://doi.org/10.1109/MECBME.2018.8402428
Wahid MF, Tafreshi R, Al-Sowaidi M, Langari R. An efficient approach to recognize hand gestures using machine-learning algorithms. In 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018. Vol. 2018-March. IEEE Computer Society. 2018. p. 171-176 https://doi.org/10.1109/MECBME.2018.8402428
Wahid, Md Ferdous ; Tafreshi, Reza ; Al-Sowaidi, Mubarak ; Langari, Reza. / An efficient approach to recognize hand gestures using machine-learning algorithms. 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018. Vol. 2018-March IEEE Computer Society, 2018. pp. 171-176
@inproceedings{a3dbaf9a866740219c99fcc902e3dbe3,
title = "An efficient approach to recognize hand gestures using machine-learning algorithms",
abstract = "Electromyography (EMG) from a subject's upper limb can be used to train a machine-learning algorithm to classify different hand gestures. However, variability in the EMG signal due to between-subject differences can substantially degrade the machine-learning performance. This variation is usually due to the differences in both anatomical and physiological properties of the muscles, levels of muscle contraction, and inherent noises from the sensors. The aim of this study is to develop a subject-independent algorithm that can accurately classify different hand gestures. To minimize the between-subject differences, some selected time-domain EMG features are normalized to the area under the averaged root mean square curve (AUC-RMS). Five adult subjects with ages ranging 20-37 years performed three hand gestures including fist, wave-in, and wave-out for ten to twelve times each. Five machine-learning algorithms, including k-nearest neighbor (KNN), discriminant analysis (DA), Na{\"i}ve Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM) were used to classify the three different hand gestures. The results showed that the EMG features were moderately to strongly correlated with the AUC-RMS values. The SVM yielded maximum classification accuracy using the original EMG features (97.56{\%}) which was significantly improved by using the normalized EMG features (98.73{\%}) (p<0.05). The accuracy distribution of all classifiers were found to be closer to mean values when using the normalized EMG features compared to using the original EMG features. The developed approach of classifying different hand gestures will be useful in biomedical applications such as controlling exoskeletons and in certain human-computer interaction settings.",
keywords = "electromyography, hand gestures, machine learning, pattern recognition",
author = "Wahid, {Md Ferdous} and Reza Tafreshi and Mubarak Al-Sowaidi and Reza Langari",
year = "2018",
month = "7",
day = "3",
doi = "10.1109/MECBME.2018.8402428",
language = "English",
volume = "2018-March",
pages = "171--176",
booktitle = "4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - An efficient approach to recognize hand gestures using machine-learning algorithms

AU - Wahid, Md Ferdous

AU - Tafreshi, Reza

AU - Al-Sowaidi, Mubarak

AU - Langari, Reza

PY - 2018/7/3

Y1 - 2018/7/3

N2 - Electromyography (EMG) from a subject's upper limb can be used to train a machine-learning algorithm to classify different hand gestures. However, variability in the EMG signal due to between-subject differences can substantially degrade the machine-learning performance. This variation is usually due to the differences in both anatomical and physiological properties of the muscles, levels of muscle contraction, and inherent noises from the sensors. The aim of this study is to develop a subject-independent algorithm that can accurately classify different hand gestures. To minimize the between-subject differences, some selected time-domain EMG features are normalized to the area under the averaged root mean square curve (AUC-RMS). Five adult subjects with ages ranging 20-37 years performed three hand gestures including fist, wave-in, and wave-out for ten to twelve times each. Five machine-learning algorithms, including k-nearest neighbor (KNN), discriminant analysis (DA), Naïve Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM) were used to classify the three different hand gestures. The results showed that the EMG features were moderately to strongly correlated with the AUC-RMS values. The SVM yielded maximum classification accuracy using the original EMG features (97.56%) which was significantly improved by using the normalized EMG features (98.73%) (p<0.05). The accuracy distribution of all classifiers were found to be closer to mean values when using the normalized EMG features compared to using the original EMG features. The developed approach of classifying different hand gestures will be useful in biomedical applications such as controlling exoskeletons and in certain human-computer interaction settings.

AB - Electromyography (EMG) from a subject's upper limb can be used to train a machine-learning algorithm to classify different hand gestures. However, variability in the EMG signal due to between-subject differences can substantially degrade the machine-learning performance. This variation is usually due to the differences in both anatomical and physiological properties of the muscles, levels of muscle contraction, and inherent noises from the sensors. The aim of this study is to develop a subject-independent algorithm that can accurately classify different hand gestures. To minimize the between-subject differences, some selected time-domain EMG features are normalized to the area under the averaged root mean square curve (AUC-RMS). Five adult subjects with ages ranging 20-37 years performed three hand gestures including fist, wave-in, and wave-out for ten to twelve times each. Five machine-learning algorithms, including k-nearest neighbor (KNN), discriminant analysis (DA), Naïve Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM) were used to classify the three different hand gestures. The results showed that the EMG features were moderately to strongly correlated with the AUC-RMS values. The SVM yielded maximum classification accuracy using the original EMG features (97.56%) which was significantly improved by using the normalized EMG features (98.73%) (p<0.05). The accuracy distribution of all classifiers were found to be closer to mean values when using the normalized EMG features compared to using the original EMG features. The developed approach of classifying different hand gestures will be useful in biomedical applications such as controlling exoskeletons and in certain human-computer interaction settings.

KW - electromyography

KW - hand gestures

KW - machine learning

KW - pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=85050019917&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050019917&partnerID=8YFLogxK

U2 - 10.1109/MECBME.2018.8402428

DO - 10.1109/MECBME.2018.8402428

M3 - Conference contribution

AN - SCOPUS:85050019917

VL - 2018-March

SP - 171

EP - 176

BT - 4th IEEE Middle East Conference on Biomedical Engineering, MECBME 2018

PB - IEEE Computer Society

ER -