EMG pattern recognition using decomposition techniques for constructing multiclass classifiers

Huaiqi Huang; Tao Li; Claudio Bruschini; Christian Enz; Volker M. Koch; Jorn Justiz; Christian Antfolk

doi:10.1109/BIOROB.2016.7523810

EMG pattern recognition using decomposition techniques for constructing multiclass classifiers

Huaiqi Huang, Tao Li, Claudio Bruschini, Christian Enz, Volker M. Koch, Jorn Justiz, Christian Antfolk

Division for Biomedical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

Abstract

To improve the dexterity of multi-functional myoelectric prosthetic hand, more accurate hand gesture recognition based on surface electromyographic (sEMG) signal is needed. This paper evaluates two types of time-domain EMG features, one independent feature and one combined feature including four features. The selected features from eight subjects with 13 finger movements were tested with four decomposed multi-class support vector machines (SVM), four decomposed linear discriminant analyses (LDA) and a multi-class LDA. The classification accuracy, training, and classification time are compared. The results have shown that the combined features decrease error rate, and binary tree based decomposition multiclass classifiers yield the highest classification success rate (88.2%) with relatively low training and classification time.

Original language	English
Title of host publication	2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016
Publisher	IEEE Computer Society
Pages	1296-1301
Number of pages	6
ISBN (Electronic)	9781509032877
DOIs	https://doi.org/10.1109/BIOROB.2016.7523810
Publication status	Published - 2016 Jul 26
Event	6th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016 - Singapore, Singapore Duration: 2016 Jun 26 → 2016 Jun 29

Conference

Conference	6th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016
Country/Territory	Singapore
City	Singapore
Period	2016/06/26 → 2016/06/29

Subject classification (UKÄ)

Electrical Engineering, Electronic Engineering, Information Engineering
Other Medical and Health Sciences not elsewhere specified

Access to Document

10.1109/BIOROB.2016.7523810

Cite this

Huang, H., Li, T., Bruschini, C., Enz, C., Koch, V. M., Justiz, J., & Antfolk, C. (2016). EMG pattern recognition using decomposition techniques for constructing multiclass classifiers. In 2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016 (pp. 1296-1301). Article 7523810 IEEE Computer Society. https://doi.org/10.1109/BIOROB.2016.7523810

@inproceedings{779b71ae65a646eca11f548ec8086899,

title = "EMG pattern recognition using decomposition techniques for constructing multiclass classifiers",

abstract = "To improve the dexterity of multi-functional myoelectric prosthetic hand, more accurate hand gesture recognition based on surface electromyographic (sEMG) signal is needed. This paper evaluates two types of time-domain EMG features, one independent feature and one combined feature including four features. The selected features from eight subjects with 13 finger movements were tested with four decomposed multi-class support vector machines (SVM), four decomposed linear discriminant analyses (LDA) and a multi-class LDA. The classification accuracy, training, and classification time are compared. The results have shown that the combined features decrease error rate, and binary tree based decomposition multiclass classifiers yield the highest classification success rate (88.2%) with relatively low training and classification time.",

author = "Huaiqi Huang and Tao Li and Claudio Bruschini and Christian Enz and Koch, {Volker M.} and Jorn Justiz and Christian Antfolk",

year = "2016",

month = jul,

day = "26",

doi = "10.1109/BIOROB.2016.7523810",

language = "English",

pages = "1296--1301",

booktitle = "2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016",

publisher = "IEEE Computer Society",

address = "United States",

note = "6th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016 ; Conference date: 26-06-2016 Through 29-06-2016",

}

Huang, H, Li, T, Bruschini, C, Enz, C, Koch, VM, Justiz, J & Antfolk, C 2016, EMG pattern recognition using decomposition techniques for constructing multiclass classifiers. in 2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016., 7523810, IEEE Computer Society, pp. 1296-1301, 6th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016, Singapore, Singapore, 2016/06/26. https://doi.org/10.1109/BIOROB.2016.7523810

EMG pattern recognition using decomposition techniques for constructing multiclass classifiers. / Huang, Huaiqi; Li, Tao; Bruschini, Claudio et al.
2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016. IEEE Computer Society, 2016. p. 1296-1301 7523810.

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

TY - GEN

T1 - EMG pattern recognition using decomposition techniques for constructing multiclass classifiers

AU - Huang, Huaiqi

AU - Li, Tao

AU - Bruschini, Claudio

AU - Enz, Christian

AU - Koch, Volker M.

AU - Justiz, Jorn

AU - Antfolk, Christian

PY - 2016/7/26

Y1 - 2016/7/26

N2 - To improve the dexterity of multi-functional myoelectric prosthetic hand, more accurate hand gesture recognition based on surface electromyographic (sEMG) signal is needed. This paper evaluates two types of time-domain EMG features, one independent feature and one combined feature including four features. The selected features from eight subjects with 13 finger movements were tested with four decomposed multi-class support vector machines (SVM), four decomposed linear discriminant analyses (LDA) and a multi-class LDA. The classification accuracy, training, and classification time are compared. The results have shown that the combined features decrease error rate, and binary tree based decomposition multiclass classifiers yield the highest classification success rate (88.2%) with relatively low training and classification time.

AB - To improve the dexterity of multi-functional myoelectric prosthetic hand, more accurate hand gesture recognition based on surface electromyographic (sEMG) signal is needed. This paper evaluates two types of time-domain EMG features, one independent feature and one combined feature including four features. The selected features from eight subjects with 13 finger movements were tested with four decomposed multi-class support vector machines (SVM), four decomposed linear discriminant analyses (LDA) and a multi-class LDA. The classification accuracy, training, and classification time are compared. The results have shown that the combined features decrease error rate, and binary tree based decomposition multiclass classifiers yield the highest classification success rate (88.2%) with relatively low training and classification time.

UR - http://www.scopus.com/inward/record.url?scp=84983446432&partnerID=8YFLogxK

U2 - 10.1109/BIOROB.2016.7523810

DO - 10.1109/BIOROB.2016.7523810

M3 - Paper in conference proceeding

AN - SCOPUS:84983446432

SP - 1296

EP - 1301

BT - 2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016

PB - IEEE Computer Society

T2 - 6th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2016

Y2 - 26 June 2016 through 29 June 2016

ER -

EMG pattern recognition using decomposition techniques for constructing multiclass classifiers

Abstract

Conference

Subject classification (UKÄ)

Access to Document

Other files and links

Fingerprint

Cite this