001021215 001__ 1021215
001021215 005__ 20250204113752.0
001021215 0247_ $$2doi$$a10.1088/2632-2153/ad1a4e
001021215 0247_ $$2datacite_doi$$a10.34734/FZJ-2024-00656
001021215 0247_ $$2WOS$$aWOS:001142818000001
001021215 037__ $$aFZJ-2024-00656
001021215 082__ $$a621.3
001021215 1001_ $$0P:(DE-Juel1)186834$$aGovind, Kishan$$b0$$ufzj
001021215 245__ $$aDeep learning of crystalline defects from TEM images: a solution for the problem of ‘never enough training data’
001021215 260__ $$aBristol$$bIOP Publishing$$c2024
001021215 3367_ $$2DRIVER$$aarticle
001021215 3367_ $$2DataCite$$aOutput Types/Journal article
001021215 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1705472776_24822
001021215 3367_ $$2BibTeX$$aARTICLE
001021215 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001021215 3367_ $$00$$2EndNote$$aJournal Article
001021215 520__ $$aCrystalline defects, such as line-like dislocations, play an important role for the performance and reliability of many metallic devices. Their interaction and evolution still poses a multitude of open questions to materials science and materials physics. In-situ transmission electron microscopy (TEM) experiments can provide important insights into how dislocations behave and move. The analysis of individual video frames from such experiments can provide useful insights but is limited by the capabilities of automated identification, digitization, and quantitative extraction of the dislocations as curved objects. The vast amount of data also makes manual annotation very time consuming, thereby limiting the use of deep learning (DL)-based, automated image analysis and segmentation of the dislocation microstructure. In this work, a parametric model for generating synthetic training data for segmentation of dislocations is developed. Even though domain scientists might dismiss synthetic images as artificial, our findings show that they can result in superior performance. Additionally, we propose an enhanced DL method optimized for segmenting overlapping or intersecting dislocation lines. Upon testing this framework on four distinct real datasets, we find that a model trained only on synthetic training data can also yield high-quality results on real images–even more so if the model is further fine-tuned on a few real images. Our approach demonstrates the potential of synthetic data in overcoming the limitations of manual annotation of TEM image data of dislocation microstructure, paving the way for more efficient and accurate analysis of dislocation microstructures. Last but not least, segmenting such thin, curvilinear structures is a task that is ubiquitous in many fields, which makes our method a potential candidate for other applications as well.
001021215 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001021215 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001021215 7001_ $$0P:(DE-HGF)0$$aOliveros, Daniela$$b1
001021215 7001_ $$0P:(DE-HGF)0$$aDlouhy, Antonin$$b2
001021215 7001_ $$00000-0001-5830-0434$$aLegros, Marc$$b3
001021215 7001_ $$0P:(DE-Juel1)186075$$aSandfeld, Stefan$$b4$$eCorresponding author
001021215 773__ $$0PERI:(DE-600)3017004-7$$a10.1088/2632-2153/ad1a4e$$gVol. 5, no. 1, p. 015006 -$$n1$$p015006 -$$tMachine learning: science and technology$$v5$$x2632-2153$$y2024
001021215 8564_ $$uhttps://juser.fz-juelich.de/record/1021215/files/Govind_2024_Mach._Learn.%20_Sci._Technol._5_015006.pdf$$yOpenAccess
001021215 8564_ $$uhttps://juser.fz-juelich.de/record/1021215/files/Govind_2024_Mach._Learn.%20_Sci._Technol._5_015006.gif?subformat=icon$$xicon$$yOpenAccess
001021215 8564_ $$uhttps://juser.fz-juelich.de/record/1021215/files/Govind_2024_Mach._Learn.%20_Sci._Technol._5_015006.jpg?subformat=icon-1440$$xicon-1440$$yOpenAccess
001021215 8564_ $$uhttps://juser.fz-juelich.de/record/1021215/files/Govind_2024_Mach._Learn.%20_Sci._Technol._5_015006.jpg?subformat=icon-180$$xicon-180$$yOpenAccess
001021215 8564_ $$uhttps://juser.fz-juelich.de/record/1021215/files/Govind_2024_Mach._Learn.%20_Sci._Technol._5_015006.jpg?subformat=icon-640$$xicon-640$$yOpenAccess
001021215 8767_ $$d2024-04-12$$eAPC$$jPublish and Read
001021215 909CO $$ooai:juser.fz-juelich.de:1021215$$pdnbdelivery$$popenCost$$pVDB$$pdriver$$pOpenAPC$$popen_access$$popenaire
001021215 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)186834$$aForschungszentrum Jülich$$b0$$kFZJ
001021215 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)186075$$aForschungszentrum Jülich$$b4$$kFZJ
001021215 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001021215 9141_ $$y2024
001021215 915pc $$0PC:(DE-HGF)0000$$2APC$$aAPC keys set
001021215 915pc $$0PC:(DE-HGF)0001$$2APC$$aLocal Funding
001021215 915pc $$0PC:(DE-HGF)0002$$2APC$$aDFG OA Publikationskosten
001021215 915pc $$0PC:(DE-HGF)0107$$2APC$$aTIB: IOP Publishing 2022
001021215 915pc $$0PC:(DE-HGF)0003$$2APC$$aDOAJ Journal
001021215 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001021215 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001021215 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bMACH LEARN-SCI TECHN : 2022$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2024-08-08T17:02:38Z
001021215 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2024-08-08T17:02:38Z
001021215 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2024-08-08T17:02:38Z
001021215 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)1150$$2StatID$$aDBCoverage$$bCurrent Contents - Physical, Chemical and Earth Sciences$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2025-01-01
001021215 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bMACH LEARN-SCI TECHN : 2022$$d2025-01-01
001021215 920__ $$lyes
001021215 9201_ $$0I:(DE-Juel1)IAS-9-20201008$$kIAS-9$$lMaterials Data Science and Informatics$$x0
001021215 9801_ $$aFullTexts
001021215 980__ $$ajournal
001021215 980__ $$aVDB
001021215 980__ $$aUNRESTRICTED
001021215 980__ $$aI:(DE-Juel1)IAS-9-20201008
001021215 980__ $$aAPC