Intersected EMG heatmaps and deep learning based gesture recognition

Weijie Ke; Yannan Xing; Gaetano Di Caterina; Lykourgos Petropoulakis; John Soraghan

doi:10.1145/3383972.3383982

Intersected EMG heatmaps and deep learning based gesture recognition

Weijie Ke, Yannan Xing, Gaetano Di Caterina, Lykourgos Petropoulakis, John Soraghan

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

7 Citations (Scopus)

51 Downloads (Pure)

Abstract

Hand gesture recognition in myoelectric based prosthetic devices is a key challenge to offering effective solutions to hand/lower arm amputees. A novel hand gesture recognition methodology that employs the difference of EMG energy heatmaps as the input of a specific designed deep learning neural network is presented. Experimental results using data from real amputees indicate that the proposed design achieves 94.31% as average accuracy with best accuracy rate of 98.96%. A comparison of experimental results between the proposed novel hand gesture recognition methodology and other similar approaches indicates the superior effectiveness of the new design.

Original language	English
Pages	73-78
Number of pages	6
DOIs	https://doi.org/10.1145/3383972.3383982
Publication status	Published - 17 Feb 2020
Event	12th International Conference on Machine Learning and Computing - Shenzhen, China Duration: 15 Feb 2020 → 17 Feb 2020

Conference

Conference	12th International Conference on Machine Learning and Computing
Abbreviated title	ICMLC 2020
Country/Territory	China
City	Shenzhen
Period	15/02/20 → 17/02/20

Keywords

convolutional neural network
gesture recognition
EMG
signal processing

Access to Document

10.1145/3383972.3383982

Ke-etal-ICMLC2020-Intersected-EMG-heatmaps-and-deep-learning-based-gesture-recognitionAccepted author manuscript, 484 KB

Cite this

@conference{eeea758ffcb84c498e4c4a08aa0d60a2,

title = "Intersected EMG heatmaps and deep learning based gesture recognition",

abstract = "Hand gesture recognition in myoelectric based prosthetic devices is a key challenge to offering effective solutions to hand/lower arm amputees. A novel hand gesture recognition methodology that employs the difference of EMG energy heatmaps as the input of a specific designed deep learning neural network is presented. Experimental results using data from real amputees indicate that the proposed design achieves 94.31% as average accuracy with best accuracy rate of 98.96%. A comparison of experimental results between the proposed novel hand gesture recognition methodology and other similar approaches indicates the superior effectiveness of the new design.",

keywords = "convolutional neural network, gesture recognition, EMG, signal processing",

author = "Weijie Ke and Yannan Xing and {Di Caterina}, Gaetano and Lykourgos Petropoulakis and John Soraghan",

year = "2020",

month = feb,

day = "17",

doi = "10.1145/3383972.3383982",

language = "English",

pages = "73--78",

note = "12th International Conference on Machine Learning and Computing, ICMLC 2020 ; Conference date: 15-02-2020 Through 17-02-2020",

}

TY - CONF

T1 - Intersected EMG heatmaps and deep learning based gesture recognition

AU - Ke, Weijie

AU - Xing, Yannan

AU - Di Caterina, Gaetano

AU - Petropoulakis, Lykourgos

AU - Soraghan, John

PY - 2020/2/17

Y1 - 2020/2/17

N2 - Hand gesture recognition in myoelectric based prosthetic devices is a key challenge to offering effective solutions to hand/lower arm amputees. A novel hand gesture recognition methodology that employs the difference of EMG energy heatmaps as the input of a specific designed deep learning neural network is presented. Experimental results using data from real amputees indicate that the proposed design achieves 94.31% as average accuracy with best accuracy rate of 98.96%. A comparison of experimental results between the proposed novel hand gesture recognition methodology and other similar approaches indicates the superior effectiveness of the new design.

AB - Hand gesture recognition in myoelectric based prosthetic devices is a key challenge to offering effective solutions to hand/lower arm amputees. A novel hand gesture recognition methodology that employs the difference of EMG energy heatmaps as the input of a specific designed deep learning neural network is presented. Experimental results using data from real amputees indicate that the proposed design achieves 94.31% as average accuracy with best accuracy rate of 98.96%. A comparison of experimental results between the proposed novel hand gesture recognition methodology and other similar approaches indicates the superior effectiveness of the new design.

KW - convolutional neural network

KW - gesture recognition

KW - EMG

KW - signal processing

U2 - 10.1145/3383972.3383982

DO - 10.1145/3383972.3383982

M3 - Paper

SP - 73

EP - 78

T2 - 12th International Conference on Machine Learning and Computing

Y2 - 15 February 2020 through 17 February 2020

ER -

Intersected EMG heatmaps and deep learning based gesture recognition

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this