No Cover Image

E-Thesis 465 views

Artificial Intelligence & the steel industry: Developing an automated quality inspection system / ANASTASIA PARAMORE

Swansea University Author: ANASTASIA PARAMORE

  • Redacted version - open access under embargo until: 8th September 2024

DOI (Published version): 10.23889/SUthesis.59112

Abstract

This thesis explores possible improvements to the existing defect detection sys-tems used by Tata Steel Europe in South Wales. Exploring the possibilities of ac-curacy increases and further uses for in-house surface quality systems presents Tata Steel Europe with methods and considerations on how to...

Full description

Published: Swansea 2022
Institution: Swansea University
Degree level: Doctoral
Degree name: EngD
Supervisor: Xie, Xianghua ; Tappenden, Andrew C. ; Lewis, Simon G.
URI: https://cronfa.swan.ac.uk/Record/cronfa59112
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2022-01-07T12:04:13Z
last_indexed 2022-01-08T04:28:07Z
id cronfa59112
recordtype RisThesis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2022-01-07T12:33:09.2915816</datestamp><bib-version>v2</bib-version><id>59112</id><entry>2022-01-07</entry><title>Artificial Intelligence &amp; the steel industry: Developing an automated quality inspection system</title><swanseaauthors><author><sid>40569a88c233378e20628367d22ce87a</sid><firstname>ANASTASIA</firstname><surname>PARAMORE</surname><name>ANASTASIA PARAMORE</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2022-01-07</date><abstract>This thesis explores possible improvements to the existing defect detection sys-tems used by Tata Steel Europe in South Wales. Exploring the possibilities of ac-curacy increases and further uses for in-house surface quality systems presents Tata Steel Europe with methods and considerations on how to improve their current surface defect management tools. This research considers three dif-ferent aspects of the defect management process and tests alternatives to the systems either currently used or not yet in place.The &#xFB01;rst section considers an ensemble object detection method using Gabor &#xFB01;lters, histograms, and a random forest classi&#xFB01;er. The algorithm was applied originally as an ensemble method and then as a case-study it was split into two combinations of the methods to determine what level of ensemble complexity was optimal. The images used were a set of various types of steel defect im-ages provided by Tata Steel Europe. The ensemble method achieved acceptable results compared to the existing systems, but the medium-complexity method was optimal regarding overall accuracy, false negative rate, and speed. The sec-ond section used a set of weld hole images which were split into three quality grades by de&#xFB01;ning characteristics of the feature. These images were used to test three different neural networks, an R-CNN version of GoogLeNet, and Faster R-CNN versions of ResNet-50 and ResNet-101. These networks were tasked with classifying the quality of weld holes in steel. Accurately detecting weld holes is vital in steel production as certain production processes require speed changes for welds. As the hole punches degrade over time, so does the quality of the weld hole. When weld holes are of poor quality, they can be missed or wrongly detected. All three networks detected the weld holes very well, but classifying the quality grade was approximately 60% accurate for both ResNets and 79% accurate for the GoogLeNet. These tests highlighted the importance of data quantity and quality, including lack of bias in data. The third section looks at how colour &#xFB01;lters and greyscale methods can affect the images used for detection and classi&#xFB01;cation. An investigation into coloured light sources has not been fully explored at Tata Steel in South Wales before. Having worked with the image data for sections one and two, it highlighted how important the quality of these images is. Steel defect samples were supplied, alongside their lab-con&#xFB01;rmed defect label. Scans were taken of clean and oiled steel samples with different coloured &#xFB01;lters and a variety of common greyscale methods were used to turn these images from RGB colour to greyscale images. The greyscale values of defect to clean steel were calculated for a speci&#xFB01;c region of interest on each steel sample. This value was used as a measure of contrast between clean and defect steel. The yellow &#xFB01;lter with the Decolorize method produced the highest contrast image, higher than using no &#xFB01;lter at all. For oiled sam-ples, using no &#xFB01;lter with the Decolorize method produced the best contrast and the orange &#xFB01;lter with the Decolorize method produced the best contrast out of the &#xFB01;ltered images. The greyscale methods used had signi&#xFB01;cant effect on the contrast of the image.The signi&#xFB01;cance of this thesis is that it informs of the dif&#xFB01;culties in developing surface inspection of steel defects due to several factors, such as environment and variation of defect shape, size, colour, and frequency. The work undertaken in this study has highlighted the need for a larger pilot line to further research both the capturing of the perfect image, and &#xFB01;nding the most accurate detection and classi&#xFB01;cation methods for each grade of steel.</abstract><type>E-Thesis</type><journal/><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication>Swansea</placeOfPublication><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords>Steel defects, quality inspection systems, machine learning</keywords><publishedDay>7</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2022</publishedYear><publishedDate>2022-01-07</publishedDate><doi>10.23889/SUthesis.59112</doi><url/><notes>A selection of third party content is redacted or is partially redacted from this thesis due to copyright restrictions.ORCiD identifier: https://orcid.org/0000-0002-2328-6123</notes><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><supervisor>Xie, Xianghua ; Tappenden, Andrew C. ; Lewis, Simon G.</supervisor><degreelevel>Doctoral</degreelevel><degreename>EngD</degreename><degreesponsorsfunders>M2A funding from the European Social Fund (ESF) through the European Union&#x2019;s Convergence programme administered by the Welsh Government (c80816) and Tata Steel UK; Research grant number: c80816</degreesponsorsfunders><apcterm/><lastEdited>2022-01-07T12:33:09.2915816</lastEdited><Created>2022-01-07T11:59:07.4312033</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Engineering and Applied Sciences - Uncategorised</level></path><authors><author><firstname>ANASTASIA</firstname><surname>PARAMORE</surname><order>1</order></author></authors><documents><document><filename>Under embargo</filename><originalFilename>Under embargo</originalFilename><uploaded>2022-01-07T12:30:32.8679469</uploaded><type>Output</type><contentLength>6758184</contentLength><contentType>application/pdf</contentType><version>Redacted version - open access</version><cronfaStatus>true</cronfaStatus><embargoDate>2024-09-08T00:00:00.0000000</embargoDate><documentNotes>Copyright: The author, Anastasia Paramore, 2021.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2022-01-07T12:33:09.2915816 v2 59112 2022-01-07 Artificial Intelligence & the steel industry: Developing an automated quality inspection system 40569a88c233378e20628367d22ce87a ANASTASIA PARAMORE ANASTASIA PARAMORE true false 2022-01-07 This thesis explores possible improvements to the existing defect detection sys-tems used by Tata Steel Europe in South Wales. Exploring the possibilities of ac-curacy increases and further uses for in-house surface quality systems presents Tata Steel Europe with methods and considerations on how to improve their current surface defect management tools. This research considers three dif-ferent aspects of the defect management process and tests alternatives to the systems either currently used or not yet in place.The first section considers an ensemble object detection method using Gabor filters, histograms, and a random forest classifier. The algorithm was applied originally as an ensemble method and then as a case-study it was split into two combinations of the methods to determine what level of ensemble complexity was optimal. The images used were a set of various types of steel defect im-ages provided by Tata Steel Europe. The ensemble method achieved acceptable results compared to the existing systems, but the medium-complexity method was optimal regarding overall accuracy, false negative rate, and speed. The sec-ond section used a set of weld hole images which were split into three quality grades by defining characteristics of the feature. These images were used to test three different neural networks, an R-CNN version of GoogLeNet, and Faster R-CNN versions of ResNet-50 and ResNet-101. These networks were tasked with classifying the quality of weld holes in steel. Accurately detecting weld holes is vital in steel production as certain production processes require speed changes for welds. As the hole punches degrade over time, so does the quality of the weld hole. When weld holes are of poor quality, they can be missed or wrongly detected. All three networks detected the weld holes very well, but classifying the quality grade was approximately 60% accurate for both ResNets and 79% accurate for the GoogLeNet. These tests highlighted the importance of data quantity and quality, including lack of bias in data. The third section looks at how colour filters and greyscale methods can affect the images used for detection and classification. An investigation into coloured light sources has not been fully explored at Tata Steel in South Wales before. Having worked with the image data for sections one and two, it highlighted how important the quality of these images is. Steel defect samples were supplied, alongside their lab-confirmed defect label. Scans were taken of clean and oiled steel samples with different coloured filters and a variety of common greyscale methods were used to turn these images from RGB colour to greyscale images. The greyscale values of defect to clean steel were calculated for a specific region of interest on each steel sample. This value was used as a measure of contrast between clean and defect steel. The yellow filter with the Decolorize method produced the highest contrast image, higher than using no filter at all. For oiled sam-ples, using no filter with the Decolorize method produced the best contrast and the orange filter with the Decolorize method produced the best contrast out of the filtered images. The greyscale methods used had significant effect on the contrast of the image.The significance of this thesis is that it informs of the difficulties in developing surface inspection of steel defects due to several factors, such as environment and variation of defect shape, size, colour, and frequency. The work undertaken in this study has highlighted the need for a larger pilot line to further research both the capturing of the perfect image, and finding the most accurate detection and classification methods for each grade of steel. E-Thesis Swansea Steel defects, quality inspection systems, machine learning 7 1 2022 2022-01-07 10.23889/SUthesis.59112 A selection of third party content is redacted or is partially redacted from this thesis due to copyright restrictions.ORCiD identifier: https://orcid.org/0000-0002-2328-6123 COLLEGE NANME COLLEGE CODE Swansea University Xie, Xianghua ; Tappenden, Andrew C. ; Lewis, Simon G. Doctoral EngD M2A funding from the European Social Fund (ESF) through the European Union’s Convergence programme administered by the Welsh Government (c80816) and Tata Steel UK; Research grant number: c80816 2022-01-07T12:33:09.2915816 2022-01-07T11:59:07.4312033 Faculty of Science and Engineering School of Engineering and Applied Sciences - Uncategorised ANASTASIA PARAMORE 1 Under embargo Under embargo 2022-01-07T12:30:32.8679469 Output 6758184 application/pdf Redacted version - open access true 2024-09-08T00:00:00.0000000 Copyright: The author, Anastasia Paramore, 2021. true eng
title Artificial Intelligence & the steel industry: Developing an automated quality inspection system
spellingShingle Artificial Intelligence & the steel industry: Developing an automated quality inspection system
ANASTASIA PARAMORE
title_short Artificial Intelligence & the steel industry: Developing an automated quality inspection system
title_full Artificial Intelligence & the steel industry: Developing an automated quality inspection system
title_fullStr Artificial Intelligence & the steel industry: Developing an automated quality inspection system
title_full_unstemmed Artificial Intelligence & the steel industry: Developing an automated quality inspection system
title_sort Artificial Intelligence & the steel industry: Developing an automated quality inspection system
author_id_str_mv 40569a88c233378e20628367d22ce87a
author_id_fullname_str_mv 40569a88c233378e20628367d22ce87a_***_ANASTASIA PARAMORE
author ANASTASIA PARAMORE
author2 ANASTASIA PARAMORE
format E-Thesis
publishDate 2022
institution Swansea University
doi_str_mv 10.23889/SUthesis.59112
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Engineering and Applied Sciences - Uncategorised{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Engineering and Applied Sciences - Uncategorised
document_store_str 0
active_str 0
description This thesis explores possible improvements to the existing defect detection sys-tems used by Tata Steel Europe in South Wales. Exploring the possibilities of ac-curacy increases and further uses for in-house surface quality systems presents Tata Steel Europe with methods and considerations on how to improve their current surface defect management tools. This research considers three dif-ferent aspects of the defect management process and tests alternatives to the systems either currently used or not yet in place.The first section considers an ensemble object detection method using Gabor filters, histograms, and a random forest classifier. The algorithm was applied originally as an ensemble method and then as a case-study it was split into two combinations of the methods to determine what level of ensemble complexity was optimal. The images used were a set of various types of steel defect im-ages provided by Tata Steel Europe. The ensemble method achieved acceptable results compared to the existing systems, but the medium-complexity method was optimal regarding overall accuracy, false negative rate, and speed. The sec-ond section used a set of weld hole images which were split into three quality grades by defining characteristics of the feature. These images were used to test three different neural networks, an R-CNN version of GoogLeNet, and Faster R-CNN versions of ResNet-50 and ResNet-101. These networks were tasked with classifying the quality of weld holes in steel. Accurately detecting weld holes is vital in steel production as certain production processes require speed changes for welds. As the hole punches degrade over time, so does the quality of the weld hole. When weld holes are of poor quality, they can be missed or wrongly detected. All three networks detected the weld holes very well, but classifying the quality grade was approximately 60% accurate for both ResNets and 79% accurate for the GoogLeNet. These tests highlighted the importance of data quantity and quality, including lack of bias in data. The third section looks at how colour filters and greyscale methods can affect the images used for detection and classification. An investigation into coloured light sources has not been fully explored at Tata Steel in South Wales before. Having worked with the image data for sections one and two, it highlighted how important the quality of these images is. Steel defect samples were supplied, alongside their lab-confirmed defect label. Scans were taken of clean and oiled steel samples with different coloured filters and a variety of common greyscale methods were used to turn these images from RGB colour to greyscale images. The greyscale values of defect to clean steel were calculated for a specific region of interest on each steel sample. This value was used as a measure of contrast between clean and defect steel. The yellow filter with the Decolorize method produced the highest contrast image, higher than using no filter at all. For oiled sam-ples, using no filter with the Decolorize method produced the best contrast and the orange filter with the Decolorize method produced the best contrast out of the filtered images. The greyscale methods used had significant effect on the contrast of the image.The significance of this thesis is that it informs of the difficulties in developing surface inspection of steel defects due to several factors, such as environment and variation of defect shape, size, colour, and frequency. The work undertaken in this study has highlighted the need for a larger pilot line to further research both the capturing of the perfect image, and finding the most accurate detection and classification methods for each grade of steel.
published_date 2022-01-07T04:16:10Z
_version_ 1763754083824435200
score 11.016235

Similar Items