ASL Fingerspelling Classification for use in Robot Control

Kevin McCready; Sonya Coleman; Dermot Kerr; Emmett Kerr

doi:10.3390/engproc2024065012

ASL Fingerspelling Classification for use in Robot Control

Kevin McCready, Sonya Coleman, Dermot Kerr, Emmett Kerr

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

20 Downloads (Pure)

Abstract

This paper proposes a gesture based control system for industrial robots. To achieve
that goal, the performance of an image classifier trained on 3 different American Sign Language (ASL) fingerspelling image datasets is considered. Then, the three are combined into a single larger dataset, and the classifier trained on that. The results of this process is then compared with the original three.

Original language	English
Title of host publication	Engineering Proceedings
Publisher	MDPI
Pages	31-32
Number of pages	2
Volume	65
Edition	1
DOIs	https://doi.org/10.3390/engproc2024065012
Publication status	Published (in print/issue) - 2024
Event	The 39th International Manufacturing Conference: Smart Manufacturing - The Next Generation - Ulster University, Magee Campus, Derry/Londonderry, Northern Ireland Duration: 24 Aug 2023 → 25 Aug 2023 https://www.manufacturingcouncil.ie/imc39-2023

Publication series

Name	Engineering Proceedings
Publisher	MDPI
ISSN (Print)	2673-4591

Conference

Conference	The 39th International Manufacturing Conference
Abbreviated title	IMC39 2023
Country/Territory	Northern Ireland
City	Derry/Londonderry
Period	24/08/23 → 25/08/23
Internet address	https://www.manufacturingcouncil.ie/imc39-2023

Bibliographical note

Publisher Copyright:
© 2024 by the authors.

Data Access Statement

The data presented in this study are openly available in American Sign Language Dataset at: https://www.kaggle.com/datasets/ayuraj/asl-dataset, American Sign Language at 10.34740/kaggle/dsv/2184214, and ASL Alphabet at 10.34740/kaggle/dsv/29550.

Keywords

Sign language
Machine Vision
Convolutional Neural Networks
Visual Communication
sign language
visual communication
convolutional neural networks
machine vision

Access to Document

10.3390/engproc2024065012

engproc-65-00012 (1)Final published version, 447 KBLicence: CC BY

Cite this

@inproceedings{58dc97b726ad4db682acc960ba7b3fca,

title = "ASL Fingerspelling Classification for use in Robot Control",

abstract = "This paper proposes a gesture based control system for industrial robots. To achievethat goal, the performance of an image classifier trained on 3 different American Sign Language (ASL) fingerspelling image datasets is considered. Then, the three are combined into a single larger dataset, and the classifier trained on that. The results of this process is then compared with the original three.",

keywords = "Sign language, Machine Vision, Convolutional Neural Networks, Visual Communication, sign language, visual communication, convolutional neural networks, machine vision",

author = "Kevin McCready and Sonya Coleman and Dermot Kerr and Emmett Kerr",

note = "Publisher Copyright: {\textcopyright} 2024 by the authors.; The 39th International Manufacturing Conference : Smart Manufacturing - The Next Generation, IMC39 2023 ; Conference date: 24-08-2023 Through 25-08-2023",

year = "2024",

doi = "10.3390/engproc2024065012",

language = "English",

volume = "65",

series = "Engineering Proceedings",

publisher = "MDPI",

pages = "31--32",

booktitle = "Engineering Proceedings",

address = "Switzerland",

edition = "1",

url = "https://www.manufacturingcouncil.ie/imc39-2023",

}

TY - GEN

T1 - ASL Fingerspelling Classification for use in Robot Control

AU - McCready, Kevin

AU - Coleman, Sonya

AU - Kerr, Dermot

AU - Kerr, Emmett

PY - 2024

Y1 - 2024

N2 - This paper proposes a gesture based control system for industrial robots. To achievethat goal, the performance of an image classifier trained on 3 different American Sign Language (ASL) fingerspelling image datasets is considered. Then, the three are combined into a single larger dataset, and the classifier trained on that. The results of this process is then compared with the original three.

AB - This paper proposes a gesture based control system for industrial robots. To achievethat goal, the performance of an image classifier trained on 3 different American Sign Language (ASL) fingerspelling image datasets is considered. Then, the three are combined into a single larger dataset, and the classifier trained on that. The results of this process is then compared with the original three.

KW - Sign language

KW - Machine Vision

KW - Convolutional Neural Networks

KW - Visual Communication

KW - sign language

KW - visual communication

KW - convolutional neural networks

KW - machine vision

UR - https://www.manufacturingcouncil.ie/imc39-2023/programme-and-proceedings

UR - http://www.scopus.com/inward/record.url?scp=85196498350&partnerID=8YFLogxK

U2 - 10.3390/engproc2024065012

DO - 10.3390/engproc2024065012

M3 - Conference contribution

VL - 65

T3 - Engineering Proceedings

SP - 31

EP - 32

BT - Engineering Proceedings

PB - MDPI

T2 - The 39th International Manufacturing Conference

Y2 - 24 August 2023 through 25 August 2023

ER -

ASL Fingerspelling Classification for use in Robot Control

Abstract

Publication series

Conference

Bibliographical note

Data Access Statement

Keywords

Access to Document

Other files and links

Fingerprint

Cite this