Publication:
Microstructure quality control of steels using deep learning

Simple item page
cris.virtual.departmentFraunhofer-Institut für Werkstoffmechanik IWM
cris.virtual.departmentFraunhofer-Institut für Werkstoffmechanik IWM
cris.virtual.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtualsource.departmentf1c22492-b2da-429a-b625-a618154dbc9d
cris.virtualsource.department9b49b6ab-7231-4bc7-89e7-49ef02e35d74
cris.virtualsource.orcidf1c22492-b2da-429a-b625-a618154dbc9d
cris.virtualsource.orcid9b49b6ab-7231-4bc7-89e7-49ef02e35d74
crisou.acronymIWM
dc.contributor.authorDurmaz, Ali Riza
dc.contributor.authorPotu, Sai Teja
dc.contributor.authorRomich, Daniel
dc.contributor.authorMöller, Johannes J.
dc.contributor.authorNützel, Ralf
dc.date.accessioned2023-09-21T11:11:18Z
dc.date.available2023-09-21T11:11:18Z
dc.date.issued2023
dc.description.abstractIn quality control, microstructures are investigated rigorously to ensure structural integrity, exclude the presence of critical volume defects, and validate the formation of the target microstructure. For quenched, hierarchically-structured steels, the morphology of the bainitic and martensitic microstructures are of major concern to guarantee the reliability of the material under service conditions. Therefore, industries conduct small sample-size inspections of materials cross-sections through metallographers to validate the needle morphology of such microstructures. We demonstrate round-robin test results revealing that this visual grading is afflicted by pronounced subjectivity despite the thorough training of personnel. Instead, we propose a deep learning image classification approach that distinguishes steels based on their microstructure type and classifies their needle length alluding to the ISO 643 grain size assessment standard. This classification approach facilitates the reliable, objective, and automated classification of hierarchically structured steels. Specifically, an accuracy of 96% and roughly 91% is attained for the distinction of martensite/bainite subtypes and needle length, respectively. This is achieved on an image dataset that contains significant variance and labeling noise as it is acquired over more than 10 years from multiple plants, alloys, etchant applications, and light optical microscopes by many metallographers (raters). Interpretability analysis gives insights into the decision-making of these models and allows for estimating their generalization capability.
dc.description.volume10
dc.identifierhttp://dx.doi.org/10.24406/fordatis/258
dc.identifier.doi10.3389/fmats.2023.1222456
dc.identifier.urihttps://publica.fraunhofer.de/handle/publica/450891
dc.language.isoen
dc.relation.datasethttps://fordatis.fraunhofer.de/handle/fordatis/321
dc.relation.ispartofFrontiers in Materials
dc.relation.issn2296-8016
dc.subjectbainite
dc.subjectdeep learning
dc.subjectgrain size
dc.subjectmartensite
dc.subjectmicrostructure
dc.subjectquality control
dc.subjectsteel
dc.titleMicrostructure quality control of steels using deep learning
dc.typejournal article
dcterms.bibliographicCitation.articlenumber1222456
dspace.entity.typePublication
oaire.citation.volume10
oairecerif.author.affiliationFraunhofer-Institut für Werkstoffmechanik IWM
oairecerif.author.affiliationFraunhofer-Institut für Werkstoffmechanik IWM
oairecerif.author.affiliationSchaeffler Technologies AG & Co. KG, Schweinfurt
oairecerif.author.affiliationSchaeffler Technologies AG & Co. KG, Schweinfurt
oairecerif.author.affiliationSchaeffler Technologies AG & Co. KG, Schweinfurt
person.identifier.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.author.alternativeaffiliationUniversity of Freiburg, Micro and Materials Mechanics
publica.author.alternativeaffiliationLeibniz University Hannover, Institute of Mechanics and Computational Mechanics
publica.author.alternativeaffiliation#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.author.alternativeaffiliation#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.author.alternativeaffiliation#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.bestOA.pdfhttps://www.frontiersin.org/articles/10.3389/fmats.2023.1222456/pdf
publica.contributor.correspondingtrue
publica.contributor.corresponding#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.contributor.corresponding#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.contributor.corresponding#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.contributor.corresponding#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.date.scupdated2025-02-24
publica.description.pagecount13 S.
publica.fhg.departmentBauteilsicherheit und Leichtbau
publica.fhg.instituteFraunhofer-Institut für Werkstoffmechanik IWM
publica.fhg.institute-controller-group645c8d0c-45b3-43e6-84b4-21b2289d9572
publica.fhg.location#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.fhg.workgroupMeso- und Mikromechanik
publica.journal.oa#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.journal.publisher#PLACEHOLDER_PARENT_METADATA_VALUE#
publica.product.titleGrain size classification round-robin dataset
publica.rights.oaOpen Access
publica.rights.oaStatusgold
publica.rights.oaUnpaywallTrue
publica.rights.timestamp2025-01-28 14:43:07.759481

Files

Collections

Artikel