Welding defect detection with image processing on a custom small dataset: A comparative study

Abstract This work focuses on detecting defects in welding seams using the most advanced You Only Look Once (YOLO) algorithms and transfer learning. To this end, the authors prepared a small dataset of images using manual welding and compared the performance of the YOLO v5, v6, v7, and v8 methods af...

Full description

Saved in:
Bibliographic Details
Main Authors: József Szőlősi, Béla J. Szekeres, Péter Magyar, Bán Adrián, Gábor Farkas, Mátyás Andó
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:IET Collaborative Intelligent Manufacturing
Subjects:
Online Access:https://doi.org/10.1049/cim2.70005
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850085002306387968
author József Szőlősi
Béla J. Szekeres
Péter Magyar
Bán Adrián
Gábor Farkas
Mátyás Andó
author_facet József Szőlősi
Béla J. Szekeres
Péter Magyar
Bán Adrián
Gábor Farkas
Mátyás Andó
author_sort József Szőlősi
collection DOAJ
description Abstract This work focuses on detecting defects in welding seams using the most advanced You Only Look Once (YOLO) algorithms and transfer learning. To this end, the authors prepared a small dataset of images using manual welding and compared the performance of the YOLO v5, v6, v7, and v8 methods after two‐step training. Key findings reveal that YOLOv7 demonstrates superior performance, suggesting its potential as a valuable tool in automated welding quality control. The authors’ research underscores the importance of model selection. It lays the groundwork for future exploration in larger datasets and varied welding scenarios, potentially contributing to defect detection practices in manufacturing industries. The dataset and the code repository links are also provided to support our findings.
format Article
id doaj-art-4e2b940b491b4bbc8a521e3810ccac91
institution DOAJ
issn 2516-8398
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series IET Collaborative Intelligent Manufacturing
spelling doaj-art-4e2b940b491b4bbc8a521e3810ccac912025-08-20T02:43:50ZengWileyIET Collaborative Intelligent Manufacturing2516-83982024-12-0164n/an/a10.1049/cim2.70005Welding defect detection with image processing on a custom small dataset: A comparative studyJózsef Szőlősi0Béla J. Szekeres1Péter Magyar2Bán Adrián3Gábor Farkas4Mátyás Andó5ELTE Eötvös Loránd University Doctoral School of Informatics Budapest HungaryDepartment of Numerical Analysis ELTE Eötvös Loránd University Faculty of Informatics Budapest HungaryELTE Eötvös Loránd University Doctoral School of Informatics Budapest HungaryELTE Eötvös Loránd University Faculty of Informatics Szombathely HungaryDepartment of Computer Algebra ELTE Eötvös Loránd University Faculty of Informatics Budapest HungaryELTE Eötvös Loránd University Faculty of Informatics Institute of Computer Science Budapest HungaryAbstract This work focuses on detecting defects in welding seams using the most advanced You Only Look Once (YOLO) algorithms and transfer learning. To this end, the authors prepared a small dataset of images using manual welding and compared the performance of the YOLO v5, v6, v7, and v8 methods after two‐step training. Key findings reveal that YOLOv7 demonstrates superior performance, suggesting its potential as a valuable tool in automated welding quality control. The authors’ research underscores the importance of model selection. It lays the groundwork for future exploration in larger datasets and varied welding scenarios, potentially contributing to defect detection practices in manufacturing industries. The dataset and the code repository links are also provided to support our findings.https://doi.org/10.1049/cim2.70005data analysisdecision makingintelligent manufacturing systemslearning (artificial intelligence)manufacturing systemsneural nets
spellingShingle József Szőlősi
Béla J. Szekeres
Péter Magyar
Bán Adrián
Gábor Farkas
Mátyás Andó
Welding defect detection with image processing on a custom small dataset: A comparative study
IET Collaborative Intelligent Manufacturing
data analysis
decision making
intelligent manufacturing systems
learning (artificial intelligence)
manufacturing systems
neural nets
title Welding defect detection with image processing on a custom small dataset: A comparative study
title_full Welding defect detection with image processing on a custom small dataset: A comparative study
title_fullStr Welding defect detection with image processing on a custom small dataset: A comparative study
title_full_unstemmed Welding defect detection with image processing on a custom small dataset: A comparative study
title_short Welding defect detection with image processing on a custom small dataset: A comparative study
title_sort welding defect detection with image processing on a custom small dataset a comparative study
topic data analysis
decision making
intelligent manufacturing systems
learning (artificial intelligence)
manufacturing systems
neural nets
url https://doi.org/10.1049/cim2.70005
work_keys_str_mv AT jozsefszolosi weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy
AT belajszekeres weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy
AT petermagyar weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy
AT banadrian weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy
AT gaborfarkas weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy
AT matyasando weldingdefectdetectionwithimageprocessingonacustomsmalldatasetacomparativestudy