Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability

This paper presents the development of a novel real-time monitoring and detection system designed to identify the presence of welding helmets on workers’ faces during welding activities. The system employs a Convolutional Neural Network (CNN) based on the YOLOv3 algorithm and is trained a...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohammad Z. Shanti, Chan Yeob Yeun, Chung-Suk Cho, Ernesto Damiani, Tae-Yeon Kim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10818429/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850285053640179712
author Mohammad Z. Shanti
Chan Yeob Yeun
Chung-Suk Cho
Ernesto Damiani
Tae-Yeon Kim
author_facet Mohammad Z. Shanti
Chan Yeob Yeun
Chung-Suk Cho
Ernesto Damiani
Tae-Yeon Kim
author_sort Mohammad Z. Shanti
collection DOAJ
description This paper presents the development of a novel real-time monitoring and detection system designed to identify the presence of welding helmets on workers&#x2019; faces during welding activities. The system employs a Convolutional Neural Network (CNN) based on the YOLOv3 algorithm and is trained and validated using a diverse dataset that includes images with varying levels of blur, grayscale images, and drone-captured photos. The model&#x2019;s effectiveness is evaluated using five key performance metrics: accuracy, precision, recall, F1 score, and the AUC-ROC curve. Additionally, the study investigates the impact of various input image sizes, batch sizes, activation functions, and the incorporation of additional convolutional layers on model performance. The results indicate that the Swish activation function, combined with a batch size of 128, an input image size of <inline-formula> <tex-math notation="LaTeX">$256\times 256$ </tex-math></inline-formula>, and the addition of one convolutional layer, yielded superior performance. The model achieved outstanding values of 98% for precision, recall, and F1 score, along with an AUC of 0.98, underscoring its accuracy and reliability in detecting welding helmets.
format Article
id doaj-art-328a2be224bd478fbe0a95f7032d4ed4
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-328a2be224bd478fbe0a95f7032d4ed42025-08-20T01:47:25ZengIEEEIEEE Access2169-35362025-01-01132187220210.1109/ACCESS.2024.352393610818429Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet AvailabilityMohammad Z. Shanti0https://orcid.org/0000-0002-6112-7967Chan Yeob Yeun1https://orcid.org/0000-0002-1398-952XChung-Suk Cho2Ernesto Damiani3https://orcid.org/0000-0002-9557-6496Tae-Yeon Kim4https://orcid.org/0000-0003-4743-6023Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Construction System Engineering, Korea Soongsil Cyber University, Seoul, South KoreaDepartment of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Civil and Environmental Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesThis paper presents the development of a novel real-time monitoring and detection system designed to identify the presence of welding helmets on workers&#x2019; faces during welding activities. The system employs a Convolutional Neural Network (CNN) based on the YOLOv3 algorithm and is trained and validated using a diverse dataset that includes images with varying levels of blur, grayscale images, and drone-captured photos. The model&#x2019;s effectiveness is evaluated using five key performance metrics: accuracy, precision, recall, F1 score, and the AUC-ROC curve. Additionally, the study investigates the impact of various input image sizes, batch sizes, activation functions, and the incorporation of additional convolutional layers on model performance. The results indicate that the Swish activation function, combined with a batch size of 128, an input image size of <inline-formula> <tex-math notation="LaTeX">$256\times 256$ </tex-math></inline-formula>, and the addition of one convolutional layer, yielded superior performance. The model achieved outstanding values of 98% for precision, recall, and F1 score, along with an AUC of 0.98, underscoring its accuracy and reliability in detecting welding helmets.https://ieeexplore.ieee.org/document/10818429/Deep learningindustrial accidentsconstruction safetyconvolution neural networkmachine learning
spellingShingle Mohammad Z. Shanti
Chan Yeob Yeun
Chung-Suk Cho
Ernesto Damiani
Tae-Yeon Kim
Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
IEEE Access
Deep learning
industrial accidents
construction safety
convolution neural network
machine learning
title Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
title_full Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
title_fullStr Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
title_full_unstemmed Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
title_short Automated Welder Safety Assurance: A YOLOv3-Based Approach for Real-Time Detection of Welding Helmet Availability
title_sort automated welder safety assurance a yolov3 based approach for real time detection of welding helmet availability
topic Deep learning
industrial accidents
construction safety
convolution neural network
machine learning
url https://ieeexplore.ieee.org/document/10818429/
work_keys_str_mv AT mohammadzshanti automatedweldersafetyassuranceayolov3basedapproachforrealtimedetectionofweldinghelmetavailability
AT chanyeobyeun automatedweldersafetyassuranceayolov3basedapproachforrealtimedetectionofweldinghelmetavailability
AT chungsukcho automatedweldersafetyassuranceayolov3basedapproachforrealtimedetectionofweldinghelmetavailability
AT ernestodamiani automatedweldersafetyassuranceayolov3basedapproachforrealtimedetectionofweldinghelmetavailability
AT taeyeonkim automatedweldersafetyassuranceayolov3basedapproachforrealtimedetectionofweldinghelmetavailability