Classification of Functional Grasps Using Hybrid CNN/LSTM Network

Christopher Millar; N Siddique; Emmett Kerr

doi:10.1007/978-981-16-6636-0_27

Classification of Functional Grasps Using Hybrid CNN/LSTM Network

Christopher Millar, N Siddique, Emmett Kerr

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

25 Downloads (Pure)

Abstract

Gestures made by a human can be classified using Electromyography (EMG) signals collected from the forearm; even with low-frequency devices. Numerous steps are required from data collection and pre-processing through to final classification. Traditionally, an important part of EMG signal classification is extracting features from the raw signal to reduce dimensionality. It is predominantly carried out manually before the signals are input into a neural network. In this research, we successfully used a CNN to extract the features automatically, and an LSTM layer was utilised to classify the gestures. This network architecture removes a step in the gesture classification process. Using the raw signals input into a CNN/LSTM hybrid increased classification when compared with an LSTM network that required features to be manually extracted from the raw signals.

Original language	English
Title of host publication	Proceedings of the International Conference on BIg Data, IoT and Machine Learning
Editors	Mohammad Shamsul Arefin, M Shamim Kaiser, Anirban Bandyopadhyay, Md Atiqur Rahman Ahad, Kanad Ray
Publisher	SPRINGER LINK
Chapter	3
Pages	345-363
Number of pages	19
Volume	95
ISBN (Electronic)	978-981-16-6636-0
ISBN (Print)	978-981-16-6635-3
DOIs	https://doi.org/10.1007/978-981-16-6636-0_27
Publication status	Published (in print/issue) - 4 Dec 2021

Publication series

Name	Lecture Notes on Data Engineering and Communications Technologies
Publisher	Springer
Volume	95
ISSN (Print)	2367-4512
ISSN (Electronic)	2367-4520

Bibliographical note

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

CNN
Gesture classification
LSTM
sEMG

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-981-16-6636-0_27

BIM21_1264_CMILLARAuthor Accepted Manuscript, 8.62 MB

Cite this

Millar, C., Siddique, N., & Kerr, E. (2021). Classification of Functional Grasps Using Hybrid CNN/LSTM Network. In M. S. Arefin, M. S. Kaiser, A. Bandyopadhyay, M. A. R. Ahad, & K. Ray (Eds.), Proceedings of the International Conference on BIg Data, IoT and Machine Learning (Vol. 95, pp. 345-363). (Lecture Notes on Data Engineering and Communications Technologies; Vol. 95). SPRINGER LINK. https://doi.org/10.1007/978-981-16-6636-0_27

Millar, Christopher ; Siddique, N ; Kerr, Emmett. / Classification of Functional Grasps Using Hybrid CNN/LSTM Network. Proceedings of the International Conference on BIg Data, IoT and Machine Learning. editor / Mohammad Shamsul Arefin ; M Shamim Kaiser ; Anirban Bandyopadhyay ; Md Atiqur Rahman Ahad ; Kanad Ray. Vol. 95 SPRINGER LINK, 2021. pp. 345-363 (Lecture Notes on Data Engineering and Communications Technologies).

@inproceedings{b6cea81d8c57439ebb5673f42e7e119c,

title = "Classification of Functional Grasps Using Hybrid CNN/LSTM Network",

abstract = "Gestures made by a human can be classified using Electromyography (EMG) signals collected from the forearm; even with low-frequency devices. Numerous steps are required from data collection and pre-processing through to final classification. Traditionally, an important part of EMG signal classification is extracting features from the raw signal to reduce dimensionality. It is predominantly carried out manually before the signals are input into a neural network. In this research, we successfully used a CNN to extract the features automatically, and an LSTM layer was utilised to classify the gestures. This network architecture removes a step in the gesture classification process. Using the raw signals input into a CNN/LSTM hybrid increased classification when compared with an LSTM network that required features to be manually extracted from the raw signals.",

keywords = "CNN, Gesture classification, LSTM, sEMG",

author = "Christopher Millar and N Siddique and Emmett Kerr",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.",

year = "2021",

month = dec,

day = "4",

doi = "10.1007/978-981-16-6636-0_27",

language = "English",

isbn = "978-981-16-6635-3",

volume = "95",

series = "Lecture Notes on Data Engineering and Communications Technologies",

publisher = "SPRINGER LINK",

pages = "345--363",

editor = "Arefin, {Mohammad Shamsul} and Kaiser, {M Shamim} and Anirban Bandyopadhyay and Ahad, {Md Atiqur Rahman} and Kanad Ray",

booktitle = "Proceedings of the International Conference on BIg Data, IoT and Machine Learning",

}

Millar, C , Siddique, N & Kerr, E 2021, Classification of Functional Grasps Using Hybrid CNN/LSTM Network. in MS Arefin, MS Kaiser, A Bandyopadhyay, MAR Ahad & K Ray (eds), Proceedings of the International Conference on BIg Data, IoT and Machine Learning. vol. 95, Lecture Notes on Data Engineering and Communications Technologies, vol. 95, SPRINGER LINK, pp. 345-363. https://doi.org/10.1007/978-981-16-6636-0_27

Classification of Functional Grasps Using Hybrid CNN/LSTM Network. / Millar, Christopher ; Siddique, N ; Kerr, Emmett.
Proceedings of the International Conference on BIg Data, IoT and Machine Learning. ed. / Mohammad Shamsul Arefin; M Shamim Kaiser; Anirban Bandyopadhyay; Md Atiqur Rahman Ahad; Kanad Ray. Vol. 95 SPRINGER LINK, 2021. p. 345-363 (Lecture Notes on Data Engineering and Communications Technologies; Vol. 95).

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

TY - GEN

T1 - Classification of Functional Grasps Using Hybrid CNN/LSTM Network

AU - Millar, Christopher

AU - Siddique, N

AU - Kerr, Emmett

PY - 2021/12/4

Y1 - 2021/12/4

N2 - Gestures made by a human can be classified using Electromyography (EMG) signals collected from the forearm; even with low-frequency devices. Numerous steps are required from data collection and pre-processing through to final classification. Traditionally, an important part of EMG signal classification is extracting features from the raw signal to reduce dimensionality. It is predominantly carried out manually before the signals are input into a neural network. In this research, we successfully used a CNN to extract the features automatically, and an LSTM layer was utilised to classify the gestures. This network architecture removes a step in the gesture classification process. Using the raw signals input into a CNN/LSTM hybrid increased classification when compared with an LSTM network that required features to be manually extracted from the raw signals.

AB - Gestures made by a human can be classified using Electromyography (EMG) signals collected from the forearm; even with low-frequency devices. Numerous steps are required from data collection and pre-processing through to final classification. Traditionally, an important part of EMG signal classification is extracting features from the raw signal to reduce dimensionality. It is predominantly carried out manually before the signals are input into a neural network. In this research, we successfully used a CNN to extract the features automatically, and an LSTM layer was utilised to classify the gestures. This network architecture removes a step in the gesture classification process. Using the raw signals input into a CNN/LSTM hybrid increased classification when compared with an LSTM network that required features to be manually extracted from the raw signals.

KW - CNN

KW - Gesture classification

KW - LSTM

KW - sEMG

UR - https://link.springer.com/book/10.1007/978-981-16-6636-0?page=2#toc

UR - http://www.scopus.com/inward/record.url?scp=85120913457&partnerID=8YFLogxK

U2 - 10.1007/978-981-16-6636-0_27

DO - 10.1007/978-981-16-6636-0_27

M3 - Conference contribution

SN - 978-981-16-6635-3

VL - 95

T3 - Lecture Notes on Data Engineering and Communications Technologies

SP - 345

EP - 363

BT - Proceedings of the International Conference on BIg Data, IoT and Machine Learning

A2 - Arefin, Mohammad Shamsul

A2 - Kaiser, M Shamim

A2 - Bandyopadhyay, Anirban

A2 - Ahad, Md Atiqur Rahman

A2 - Ray, Kanad

PB - SPRINGER LINK

ER -

Millar C , Siddique N , Kerr E. Classification of Functional Grasps Using Hybrid CNN/LSTM Network. In Arefin MS, Kaiser MS, Bandyopadhyay A, Ahad MAR, Ray K, editors, Proceedings of the International Conference on BIg Data, IoT and Machine Learning. Vol. 95. SPRINGER LINK. 2021. p. 345-363. (Lecture Notes on Data Engineering and Communications Technologies). doi: 10.1007/978-981-16-6636-0_27

Classification of Functional Grasps Using Hybrid CNN/LSTM Network

Abstract

Publication series

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

LSTM Network Classification of Dexterous Individual Finger Movements

Cite this