Automated sign language alphabet detection

Ashwin Van Der Merwe; Elie Ngomseu Mambou; Theo G. Swart

doi:10.1109/AFRICON51333.2021.9570931

Automated sign language alphabet detection

Ashwin Van Der Merwe, Elie Ngomseu Mambou, Theo G. Swart

Electrical and Electronic Engineering Science

University of Johannesburg

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

Abstract

The possibility of using sensor-based data acquisition to supply the necessary information to an Artificial Neural Network (ANN), for South African Sign Language (SASL) alphabet recognition, is investigated. To accomplish this, a data glove was designed and implemented to generate the necessary sensory information. Thereafter, training the ANN to recognize a selection of signs in the SASL alphabet was achieved. The system was able to recognize six signs with an accuracy of 96%. This result provides a foundation for a system that will enable bilateral communication between the deaf and hearing demographic to mitigate the communication hindrance.

Original language	English
Title of host publication	Proceedings of 2021 IEEE AFRICON, AFRICON 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665419840
DOIs	https://doi.org/10.1109/AFRICON51333.2021.9570931
Publication status	Published - 13 Sept 2021
Event	2021 IEEE AFRICON, AFRICON 2021 - Virtual, Arusha, Tanzania, United Republic of Duration: 13 Sept 2021 → 15 Sept 2021

Publication series

Name	IEEE AFRICON Conference
Volume	2021-September
ISSN (Print)	2153-0025
ISSN (Electronic)	2153-0033

Conference

Conference	2021 IEEE AFRICON, AFRICON 2021
Country/Territory	Tanzania, United Republic of
City	Virtual, Arusha
Period	13/09/21 → 15/09/21

Keywords

Artificial Neural Network (ANN)
Automated Sign Detection
Data Glove
Machine Learning
South African Sign Language (SASL)

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/AFRICON51333.2021.9570931

Cite this

@inproceedings{edd997413f354162a274dd4b4828dac0,

title = "Automated sign language alphabet detection",

abstract = "The possibility of using sensor-based data acquisition to supply the necessary information to an Artificial Neural Network (ANN), for South African Sign Language (SASL) alphabet recognition, is investigated. To accomplish this, a data glove was designed and implemented to generate the necessary sensory information. Thereafter, training the ANN to recognize a selection of signs in the SASL alphabet was achieved. The system was able to recognize six signs with an accuracy of 96%. This result provides a foundation for a system that will enable bilateral communication between the deaf and hearing demographic to mitigate the communication hindrance.",

keywords = "Artificial Neural Network (ANN), Automated Sign Detection, Data Glove, Machine Learning, South African Sign Language (SASL)",

author = "{Van Der Merwe}, Ashwin and {Ngomseu Mambou}, Elie and Swart, {Theo G.}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE AFRICON, AFRICON 2021 ; Conference date: 13-09-2021 Through 15-09-2021",

year = "2021",

month = sep,

day = "13",

doi = "10.1109/AFRICON51333.2021.9570931",

language = "English",

series = "IEEE AFRICON Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings of 2021 IEEE AFRICON, AFRICON 2021",

address = "United States",

}

Van Der Merwe, A, Ngomseu Mambou, E & Swart, TG 2021, Automated sign language alphabet detection. in Proceedings of 2021 IEEE AFRICON, AFRICON 2021. IEEE AFRICON Conference, vol. 2021-September, Institute of Electrical and Electronics Engineers Inc., 2021 IEEE AFRICON, AFRICON 2021, Virtual, Arusha, Tanzania, United Republic of, 13/09/21. https://doi.org/10.1109/AFRICON51333.2021.9570931

TY - GEN

T1 - Automated sign language alphabet detection

AU - Van Der Merwe, Ashwin

AU - Ngomseu Mambou, Elie

AU - Swart, Theo G.

PY - 2021/9/13

Y1 - 2021/9/13

N2 - The possibility of using sensor-based data acquisition to supply the necessary information to an Artificial Neural Network (ANN), for South African Sign Language (SASL) alphabet recognition, is investigated. To accomplish this, a data glove was designed and implemented to generate the necessary sensory information. Thereafter, training the ANN to recognize a selection of signs in the SASL alphabet was achieved. The system was able to recognize six signs with an accuracy of 96%. This result provides a foundation for a system that will enable bilateral communication between the deaf and hearing demographic to mitigate the communication hindrance.

AB - The possibility of using sensor-based data acquisition to supply the necessary information to an Artificial Neural Network (ANN), for South African Sign Language (SASL) alphabet recognition, is investigated. To accomplish this, a data glove was designed and implemented to generate the necessary sensory information. Thereafter, training the ANN to recognize a selection of signs in the SASL alphabet was achieved. The system was able to recognize six signs with an accuracy of 96%. This result provides a foundation for a system that will enable bilateral communication between the deaf and hearing demographic to mitigate the communication hindrance.

KW - Artificial Neural Network (ANN)

KW - Automated Sign Detection

KW - Data Glove

KW - Machine Learning

KW - South African Sign Language (SASL)

UR - http://www.scopus.com/inward/record.url?scp=85118466064&partnerID=8YFLogxK

U2 - 10.1109/AFRICON51333.2021.9570931

DO - 10.1109/AFRICON51333.2021.9570931

M3 - Conference contribution

AN - SCOPUS:85118466064

T3 - IEEE AFRICON Conference

BT - Proceedings of 2021 IEEE AFRICON, AFRICON 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 IEEE AFRICON, AFRICON 2021

Y2 - 13 September 2021 through 15 September 2021

ER -

Automated sign language alphabet detection

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this