3D variational autoencoder for fingerprinting microstructure volume elements

Microstructure quantification is an important step towards establishing structure–property relationships in materials. Machine learning-based image processing methods have been shown to outperform conventional image processing techniques and are increasingly applied to microstructure quantification...

Full description

Saved in:
Bibliographic Details
Published in:Computational materials science Vol. 259; p. 114145
Main Authors: White, Michael D., Atkinson, Michael D., Plowman, Adam J., Shanthraj, Pratheek
Format: Journal Article
Language:English
Published: Elsevier B.V 01.09.2025
Subjects:
ISSN:0927-0256
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Microstructure quantification is an important step towards establishing structure–property relationships in materials. Machine learning-based image processing methods have been shown to outperform conventional image processing techniques and are increasingly applied to microstructure quantification tasks. In this work, we present a 3D variational autoencoder (VAE) for encoding microstructure volume elements (VEs) comprising voxelated crystallographic orientation data. Crystal symmetries in the orientation space are accounted for by mapping to the crystallographic fundamental zone as a preprocessing step, which allows for a continuous loss function to be used and improves the training convergence rate. The VAE is then used to encode a training set of VEs with an equiaxed polycrystalline microstructure with random texture. Accurate reconstructions are achieved with a relative average misorientation error of 3×10−2, on the test dataset, for a continuous latent space with dimension 256. We show that the model generalises well to microstructures with textures, grain sizes and aspect ratios outside the training distribution. The main aim of this paper is to present the 3D VAE as an application-agnostic method for parameterising microstructure for input into downstream tasks where microstructure dependence is required. As a proof of concept, a simple surrogate model for uniaxial crystal plasticity (CP) simulations, with a fixed load path and microstructural dependence is presented. Microstructural fingerprints, obtained by encoding VEs with the trained VAE encoder, parameterise the VEs in a low-dimensional latent space and are stored alongside the volume-averaged stress response, at each strain increment. This is then used to train a fully connected neural network mapping the input fingerprint to the resulting stress response, which acts as a surrogate model for the CP simulation. The fingerprint-based surrogate model is shown to accurately predict the microstructural dependence in the CP stress response, with a mean relative error of 2.75 MPa on unseen test data. This approach offers a significant speed-up on the order of 108 for a stress–strain curve prediction, compared to running a CP simulation. [Display omitted]
AbstractList Microstructure quantification is an important step towards establishing structure–property relationships in materials. Machine learning-based image processing methods have been shown to outperform conventional image processing techniques and are increasingly applied to microstructure quantification tasks. In this work, we present a 3D variational autoencoder (VAE) for encoding microstructure volume elements (VEs) comprising voxelated crystallographic orientation data. Crystal symmetries in the orientation space are accounted for by mapping to the crystallographic fundamental zone as a preprocessing step, which allows for a continuous loss function to be used and improves the training convergence rate. The VAE is then used to encode a training set of VEs with an equiaxed polycrystalline microstructure with random texture. Accurate reconstructions are achieved with a relative average misorientation error of 3×10−2, on the test dataset, for a continuous latent space with dimension 256. We show that the model generalises well to microstructures with textures, grain sizes and aspect ratios outside the training distribution. The main aim of this paper is to present the 3D VAE as an application-agnostic method for parameterising microstructure for input into downstream tasks where microstructure dependence is required. As a proof of concept, a simple surrogate model for uniaxial crystal plasticity (CP) simulations, with a fixed load path and microstructural dependence is presented. Microstructural fingerprints, obtained by encoding VEs with the trained VAE encoder, parameterise the VEs in a low-dimensional latent space and are stored alongside the volume-averaged stress response, at each strain increment. This is then used to train a fully connected neural network mapping the input fingerprint to the resulting stress response, which acts as a surrogate model for the CP simulation. The fingerprint-based surrogate model is shown to accurately predict the microstructural dependence in the CP stress response, with a mean relative error of 2.75 MPa on unseen test data. This approach offers a significant speed-up on the order of 108 for a stress–strain curve prediction, compared to running a CP simulation. [Display omitted]
ArticleNumber 114145
Author Atkinson, Michael D.
White, Michael D.
Plowman, Adam J.
Shanthraj, Pratheek
Author_xml – sequence: 1
  givenname: Michael D.
  orcidid: 0000-0002-7730-3240
  surname: White
  fullname: White, Michael D.
  email: mike.white@ukaea.uk
– sequence: 2
  givenname: Michael D.
  orcidid: 0000-0003-4843-699X
  surname: Atkinson
  fullname: Atkinson, Michael D.
– sequence: 3
  givenname: Adam J.
  orcidid: 0000-0002-9781-7353
  surname: Plowman
  fullname: Plowman, Adam J.
– sequence: 4
  givenname: Pratheek
  orcidid: 0000-0002-6324-0306
  surname: Shanthraj
  fullname: Shanthraj, Pratheek
BookMark eNqFkMtqwzAQRbVIoUnab6h_wK4kW5G9DOkTAt20a6HHqCjYUpFkQ_--CinddjHMMNw7nLkbtPLBA0J3BDcEk939qdFhmmRO2jUUU9YQ0pGOrdAaD5TXZbO7RpuUTrioh56u0bF9qBYZncwueDlWcs4BvA4GYmVDKec_IX5F53OZqsnpGFKOs85zhGoJ4zxBBSNM4HO6QVdWjgluf_sWfTw9vh9e6uPb8-thf6x1S2iuO8OIaltDFDeKcWILo-ktBwyW8w4bNVCqyABcqU71fd_tQCvLQGImmabtFvHL3TNMimBFAZxk_BYEi3MQ4iT-ghDnIMQliOLcX5xQ8BYHURRF-ReMi6CzMMH9e-MH1gFxoA
Cites_doi 10.1016/j.actamat.2019.08.045
10.1098/rspa.2017.0274
10.1038/s41524-022-00876-7
10.1038/s41598-018-31571-7
10.1016/j.commatsci.2022.111985
10.1016/j.matdes.2021.109544
10.1016/j.actamat.2007.10.044
10.1016/j.commatsci.2018.03.074
10.1007/s11222-007-9033-z
10.1016/j.commatsci.2018.04.030
10.1007/s10851-009-0161-2
10.1038/s41563-019-0402-8
10.1088/1361-651X/ab7150
10.1007/s11661-020-06008-4
10.1016/j.cma.2020.113234
10.1016/j.actamat.2017.05.014
10.21105/joss.07164
10.1038/s41746-019-0193-y
10.1007/s11831-022-09795-8
10.1016/j.jmps.2023.105329
10.1016/j.matdes.2021.109497
10.1186/2193-9772-3-5
10.1103/PhysRevB.73.024108
10.1016/j.actamat.2023.119514
10.2481/dsj.1.19
10.1016/j.commatsci.2015.08.011
10.1016/j.ijplas.2022.103430
10.1016/j.ijplas.2018.03.009
10.1002/adem.201901044
10.1063/1.4946894
10.1016/j.cam.2009.08.077
10.1016/j.pmatsci.2018.01.005
10.1007/s11837-011-0057-7
10.1146/annurev.matsci.32.110101.155324
10.1557/mrs.2016.163
10.12688/materialsopenres.17516.1
10.1016/j.cma.2019.112594
10.1016/j.msea.2015.07.075
10.1038/s41524-022-00764-0
10.1007/BF00344251
10.1016/j.actamat.2010.04.041
10.1561/2200000056
10.1002/wics.101
ContentType Journal Article
Copyright 2025 The Authors
Copyright_xml – notice: 2025 The Authors
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.commatsci.2025.114145
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
ExternalDocumentID 10_1016_j_commatsci_2025_114145
S0927025625004884
GroupedDBID --K
--M
.DC
.~1
0R~
1B1
1~.
1~5
29F
4.4
457
4G.
5GY
5VS
6I.
7-5
71M
8P~
9JN
AABXZ
AAEDT
AAEDW
AAEPC
AAFTH
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYWO
ABFNM
ABJNI
ABMAC
ABWVN
ABXDB
ABXRA
ACDAQ
ACGFS
ACLOT
ACNNM
ACRLP
ACRPL
ACVFH
ADBBV
ADCNI
ADEZE
ADMUD
ADNMO
AEBSH
AECPX
AEIPS
AEKER
AENEX
AEUPX
AEZYN
AFJKZ
AFPUW
AFRZQ
AFTJW
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AHJVU
AI.
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
APXCP
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFKBS
EFLBG
EJD
EO8
EO9
EP2
EP3
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
HLZ
HVGLF
HZ~
IHE
J1W
JJJVA
KOM
LG9
M24
M41
MAGPM
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RNS
ROL
RPZ
SBC
SDF
SDG
SES
SEW
SMS
SPC
SPCBC
SPD
SSM
SST
SSZ
T5K
VH1
WUQ
XPP
ZMT
~G-
~HD
9DU
AAYXX
CITATION
ID FETCH-LOGICAL-c312t-4d51b33d1b7db571f414d8f7e0ef7740db922b19e7bb4b88846ecbf5ea05a5c23
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001547379800002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0927-0256
IngestDate Sat Nov 29 06:55:28 EST 2025
Sat Oct 04 17:00:13 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Microstructure characterisation
Feature learning
Crystal plasticity
Fingerprinting
Surrogate model
Machine learning
Language English
License This is an open access article under the CC BY license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c312t-4d51b33d1b7db571f414d8f7e0ef7740db922b19e7bb4b88846ecbf5ea05a5c23
ORCID 0000-0002-7730-3240
0000-0002-6324-0306
0000-0002-9781-7353
0000-0003-4843-699X
OpenAccessLink https://dx.doi.org/10.1016/j.commatsci.2025.114145
ParticipantIDs crossref_primary_10_1016_j_commatsci_2025_114145
elsevier_sciencedirect_doi_10_1016_j_commatsci_2025_114145
PublicationCentury 2000
PublicationDate September 2025
2025-09-00
PublicationDateYYYYMMDD 2025-09-01
PublicationDate_xml – month: 09
  year: 2025
  text: September 2025
PublicationDecade 2020
PublicationTitle Computational materials science
PublicationYear 2025
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Kingma, Welling (b45) 2019; 12
Mianroodi, Shanthraj, Liu, Vakili, Roongta, Siboni, Perchikov, Bai, Svendsen, Roters, Raabe, Diehl (b2) 2022; 8
DeCost, Francis, Holm (b25) 2017; 133
Van Der Giessen, Schultz, Bertin, Bulatov, Cai, Csányi, Foiles, Geers, González, Hütter, Kim, Kochmann, LLorca, Mattsson, Rottler, Shluger, Sills, Steinbach, Strachan, Tadmor (b1) 2020; 28
DeCost, Holm (b24) 2015; 110
Roters, Diehl, Shanthraj, Eisenlohr, Reuber, Wong, Maiti, Ebrahimi, Hochrainer, Fabritius, Nikolov, Friák, Fujita, Grilli, Janssens, Jia, Kok, Ma, Meier, Werner, Stricker, Weygand, Raabe (b40) 2019; 158
Huynh (b49) 2009; 35
Björk, Myhre, Johansen (b46) 2022
McDowell, LeSar (b15) 2016; 41
Burnett, Withers (b13) 2019; 18
Torquato (b18) 2002; 32
Oommen, Shukla, Goswami, Dingreville, Karniadakis (b37) 2022; 8
Otto de Mentock, Roongta, Roters, Eisenlohr, Martin (b43) 2025; 10
Hahn, Meyers (b17) 2015; 646
Michael White, . Michael Atkinson, . Adam Plowman, . Pratheek Shanthraj, 3D variational autoencoder for fingerprinting microstructure volume elements: Supplementary Data.
Fukushima (b28) 1980; 36
Liu, Ocegueda, Trautner, Stuart, Bhattacharya (b9) 2023; 178
Bostanabad, Zhang, Li, Kearney, Brinson, Apley, Liu, Chen (b16) 2018; 95
Bonatti, Berisha, Mohr (b8) 2022; 158
.
Chen, Xu, Chawla, Ren, Jiao (b23) 2019; 179
Krakow, Bennett, Johnstone, Vukmanovic, Solano-Alvarez, Lainé, Einsle, Midgley, Rae, Hielscher (b39) 2017; 473
Kalidindi, Niezgoda, Salem (b19) 2011; 63
Michael White, lightning-vae3d. URL
White, Gowtham, Jeyapriya, Basu, Withers, Race (b36) 2024
Geers, Kouznetsova, Brekelmans (b5) 2010; 234
Li, Zhang, Zhao, Burkhart, Brinson, Chen (b31) 2018; 8
Wu, Nguyen, Kilingar, Noels (b11) 2020; 369
Krizhevsky, Sutskever, Hinton (b29) 2012
Ghavamian, Simone (b10) 2019; 357
Suh, Rajagopalan, Li, Rajan (b27) 2002; 1
Kingma, Welling (b32) 2013
Paszke, Gross, Massa, Lerer, Bradbury, Chanan, Killeen, Lin, Gimelshein, Antiga, Desmaison, Kopf, Yang, DeVito, Raison, Tejani, Chilamkurthy, Steiner, Fang, Bai, Chintala (b42) 2019
URL
Alber, Buganza Tepole, Cannon, De, Dura-Bernal, Garikipati, Karniadakis, Lytton, Perdikaris, Petzold, Kuhl (b6) 2019; 2
Desai, Shrivastava, D’Elia, Najm, Dingreville (b38) 2024; 263
Fullwood, Niezgoda, Kalidindi (b21) 2008; 56
Fernandez-Zelaia, Mayeur, Cheng, Lee, Knipe, Kadau (b12) 2024
Cang, Li, Yao, Jiao, Ren (b33) 2018; 150
von Luxburg (b51) 2007; 17
White, Tarakanov, Withers, Race, Law (b26) 2023; 218
Diehl, Wang, Liu, Rezaei Mianroodi, Han, Ma, Kok, Roters, Shanthraj (b3) 2020; 22
Groeber, Jackson (b41) 2014; 3
Liu, Shanthraj, Diehl, Roters, Dong, Dong, Ding, Raabe (b48) 2018; 106
Agrawal, Choudhary (b14) 2016; 4
Niezgoda, Turner, Fullwood, Kalidindi (b20) 2010; 58
Lu, Tadmor, Kaxiras (b4) 2006; 73
He, Zhang, Ren, Sun (b35) 2015
Hu, Li, Wang, Wang (b22) 2021; 201
Holm, Cohn, Gao, Kitahara, Matson, Lei, Yarasi (b30) 2020; 51
Kim, Park, Jung, Asghari-Rad, Lee, Kim, Jung, Kim (b34) 2021; 202
Bishara, Xie, Liu, Li (b7) 2023; 30
Kingma, Ba (b44) 2017
Abdi, Williams (b50) 2010; 2
Plowman, Jedrasiak, Jailin, Crowther, Mishra, Shanthraj, Quinta Da Fonseca (b47) 2023; 2
Abdi (10.1016/j.commatsci.2025.114145_b50) 2010; 2
Kalidindi (10.1016/j.commatsci.2025.114145_b19) 2011; 63
Fukushima (10.1016/j.commatsci.2025.114145_b28) 1980; 36
He (10.1016/j.commatsci.2025.114145_b35) 2015
Hahn (10.1016/j.commatsci.2025.114145_b17) 2015; 646
Chen (10.1016/j.commatsci.2025.114145_b23) 2019; 179
Lu (10.1016/j.commatsci.2025.114145_b4) 2006; 73
McDowell (10.1016/j.commatsci.2025.114145_b15) 2016; 41
Liu (10.1016/j.commatsci.2025.114145_b48) 2018; 106
Holm (10.1016/j.commatsci.2025.114145_b30) 2020; 51
Bostanabad (10.1016/j.commatsci.2025.114145_b16) 2018; 95
Huynh (10.1016/j.commatsci.2025.114145_b49) 2009; 35
10.1016/j.commatsci.2025.114145_b52
Kim (10.1016/j.commatsci.2025.114145_b34) 2021; 202
Kingma (10.1016/j.commatsci.2025.114145_b44) 2017
Liu (10.1016/j.commatsci.2025.114145_b9) 2023; 178
Plowman (10.1016/j.commatsci.2025.114145_b47) 2023; 2
Hu (10.1016/j.commatsci.2025.114145_b22) 2021; 201
White (10.1016/j.commatsci.2025.114145_b36) 2024
White (10.1016/j.commatsci.2025.114145_b26) 2023; 218
Wu (10.1016/j.commatsci.2025.114145_b11) 2020; 369
10.1016/j.commatsci.2025.114145_b53
Burnett (10.1016/j.commatsci.2025.114145_b13) 2019; 18
Niezgoda (10.1016/j.commatsci.2025.114145_b20) 2010; 58
Oommen (10.1016/j.commatsci.2025.114145_b37) 2022; 8
Geers (10.1016/j.commatsci.2025.114145_b5) 2010; 234
DeCost (10.1016/j.commatsci.2025.114145_b24) 2015; 110
Krakow (10.1016/j.commatsci.2025.114145_b39) 2017; 473
Agrawal (10.1016/j.commatsci.2025.114145_b14) 2016; 4
DeCost (10.1016/j.commatsci.2025.114145_b25) 2017; 133
Mianroodi (10.1016/j.commatsci.2025.114145_b2) 2022; 8
Bonatti (10.1016/j.commatsci.2025.114145_b8) 2022; 158
Krizhevsky (10.1016/j.commatsci.2025.114145_b29) 2012
Bishara (10.1016/j.commatsci.2025.114145_b7) 2023; 30
Roters (10.1016/j.commatsci.2025.114145_b40) 2019; 158
Groeber (10.1016/j.commatsci.2025.114145_b41) 2014; 3
Kingma (10.1016/j.commatsci.2025.114145_b45) 2019; 12
Li (10.1016/j.commatsci.2025.114145_b31) 2018; 8
Fullwood (10.1016/j.commatsci.2025.114145_b21) 2008; 56
von Luxburg (10.1016/j.commatsci.2025.114145_b51) 2007; 17
Ghavamian (10.1016/j.commatsci.2025.114145_b10) 2019; 357
Björk (10.1016/j.commatsci.2025.114145_b46) 2022
Kingma (10.1016/j.commatsci.2025.114145_b32) 2013
Otto de Mentock (10.1016/j.commatsci.2025.114145_b43) 2025; 10
Desai (10.1016/j.commatsci.2025.114145_b38) 2024; 263
Alber (10.1016/j.commatsci.2025.114145_b6) 2019; 2
Diehl (10.1016/j.commatsci.2025.114145_b3) 2020; 22
Torquato (10.1016/j.commatsci.2025.114145_b18) 2002; 32
Suh (10.1016/j.commatsci.2025.114145_b27) 2002; 1
Fernandez-Zelaia (10.1016/j.commatsci.2025.114145_b12) 2024
Van Der Giessen (10.1016/j.commatsci.2025.114145_b1) 2020; 28
Cang (10.1016/j.commatsci.2025.114145_b33) 2018; 150
Paszke (10.1016/j.commatsci.2025.114145_b42) 2019
References_xml – volume: 150
  start-page: 212
  year: 2018
  end-page: 221
  ident: b33
  article-title: Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model
  publication-title: Comput. Mater. Sci.
– volume: 35
  start-page: 155
  year: 2009
  end-page: 164
  ident: b49
  article-title: Metrics for 3D rotations: Comparison and analysis
  publication-title: J. Math. Imaging Vision
– volume: 158
  start-page: 420
  year: 2019
  end-page: 478
  ident: b40
  article-title: DAMASK – the Düsseldorf advanced material simulation kit for modeling multi-physics crystal plasticity, thermal, and damage phenomena from the single crystal up to the component scale
  publication-title: Comput. Mater. Sci.
– volume: 369
  year: 2020
  ident: b11
  article-title: A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths
  publication-title: Comput. Methods Appl. Mech. Engrg.
– reference: . URL
– year: 2015
  ident: b35
  article-title: Deep residual learning for image recognition
– volume: 1
  start-page: 19
  year: 2002
  end-page: 26
  ident: b27
  article-title: The application of principal component analysis to materials science data
  publication-title: Data Sci. J.
– volume: 2
  start-page: 433
  year: 2010
  end-page: 459
  ident: b50
  article-title: Principal component analysis: Principal component analysis
  publication-title: Wiley Interdiscip. Rev.: Comput. Stat.
– volume: 32
  start-page: 77
  year: 2002
  end-page: 111
  ident: b18
  article-title: Statistical description of microstructures
  publication-title: Annu. Rev. Mater. Res.
– volume: 8
  start-page: 190
  year: 2022
  ident: b37
  article-title: Learning two-phase microstructure evolution using neural operators and autoencoder architectures
  publication-title: Npj Comput. Mater.
– volume: 473
  year: 2017
  ident: b39
  article-title: On three-dimensional misorientation spaces
  publication-title: Proc. R. Soc. A: Math. Phys. Eng. Sci.
– volume: 58
  start-page: 4432
  year: 2010
  end-page: 4445
  ident: b20
  article-title: Optimized structure based representative volume element sets reflecting the ensemble-averaged 2-point statistics
  publication-title: Acta Mater.
– volume: 56
  start-page: 942
  year: 2008
  end-page: 948
  ident: b21
  article-title: Microstructure reconstructions from 2-point statistics using phase-recovery algorithms
  publication-title: Acta Mater.
– volume: 4
  year: 2016
  ident: b14
  article-title: Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science
  publication-title: APL Mater.
– volume: 95
  start-page: 1
  year: 2018
  end-page: 41
  ident: b16
  article-title: Computational microstructure characterization and reconstruction: Review of the state-of-the-art techniques
  publication-title: Prog. Mater. Sci.
– volume: 218
  year: 2023
  ident: b26
  article-title: Digital fingerprinting of microstructures
  publication-title: Comput. Mater. Sci.
– volume: 2
  start-page: 2
  year: 2023
  ident: b47
  article-title: A novel integrated framework for reproducible formability predictions using virtual materials testing
  publication-title: Mater. Open Res.
– volume: 179
  start-page: 317
  year: 2019
  end-page: 327
  ident: b23
  article-title: Hierarchical n-point polytope functions for quantitative representation of complex heterogeneous materials and microstructural evolution
  publication-title: Acta Mater.
– year: 2012
  ident: b29
  article-title: ImageNet classification with deep convolutional neural networks
  publication-title: Advances in Neural Information Processing Systems
– volume: 158
  year: 2022
  ident: b8
  article-title: From CP-FFT to CP-RNN: Recurrent neural network surrogate model of crystal plasticity
  publication-title: Int. J. Plast.
– volume: 17
  start-page: 395
  year: 2007
  end-page: 416
  ident: b51
  article-title: A tutorial on spectral clustering
  publication-title: Stat. Comput.
– volume: 133
  start-page: 30
  year: 2017
  end-page: 40
  ident: b25
  article-title: Exploring the microstructure manifold: Image texture representations applied to ultrahigh carbon steel microstructures
  publication-title: Acta Mater.
– volume: 646
  start-page: 101
  year: 2015
  end-page: 134
  ident: b17
  article-title: Grain-size dependent mechanical behavior of nanocrystalline metals
  publication-title: Mater. Sci. Eng.: A
– volume: 110
  start-page: 126
  year: 2015
  end-page: 133
  ident: b24
  article-title: A computer vision approach for automated analysis and classification of microstructural image data
  publication-title: Comput. Mater. Sci.
– volume: 263
  year: 2024
  ident: b38
  article-title: Trade-offs in the latent representation of microstructure evolution
  publication-title: Acta Mater.
– volume: 178
  year: 2023
  ident: b9
  article-title: Learning macroscopic internal variables and history dependence from microscopic models
  publication-title: J. Mech. Phys. Solids
– year: 2024
  ident: b36
  article-title: Exploring descriptors for titanium microstructure via digital fingerprints from variational autoencoders
– year: 2022
  ident: b46
  article-title: Simpler is better: spectral regularization and up-sampling techniques for variational autoencoders
– reference: . Michael White, . Michael Atkinson, . Adam Plowman, . Pratheek Shanthraj, 3D variational autoencoder for fingerprinting microstructure volume elements: Supplementary Data.
– volume: 36
  start-page: 193
  year: 1980
  end-page: 202
  ident: b28
  article-title: Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position
  publication-title: Biol. Cybernet.
– year: 2024
  ident: b12
  article-title: Self-supervised feature distillation and design of experiments for efficient training of micromechanical deep learning surrogates
– start-page: 8024
  year: 2019
  end-page: 8035
  ident: b42
  article-title: PyTorch: An imperative style, high-performance deep learning library
  publication-title: Advances in Neural Information Processing Systems 32
– volume: 10
  start-page: 7164
  year: 2025
  ident: b43
  article-title: A python library for pre- and post-processing of DAMASK simulations
  publication-title: J. Open Source Softw.
– reference: . Michael White, lightning-vae3d. URL
– year: 2017
  ident: b44
  article-title: Adam: A method for stochastic optimization
– volume: 2
  start-page: 115
  year: 2019
  ident: b6
  article-title: Integrating machine learning and multiscale modeling—perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences
  publication-title: Npj Digit. Med.
– volume: 18
  start-page: 1041
  year: 2019
  end-page: 1049
  ident: b13
  article-title: Completing the picture through correlative characterization
  publication-title: Nat. Mater.
– volume: 41
  start-page: 587
  year: 2016
  end-page: 593
  ident: b15
  article-title: The need for microstructure informatics in process–structure–property relations
  publication-title: MRS Bull.
– volume: 8
  start-page: 93
  year: 2022
  ident: b2
  article-title: Modeling and simulation of microstructure in metallic systems based on multi-physics approaches
  publication-title: Npj Comput. Mater.
– volume: 3
  start-page: 56
  year: 2014
  end-page: 72
  ident: b41
  article-title: DREAM.3D: A digital representation environment for the analysis of microstructure in 3D
  publication-title: Integr. Mater. Manuf. Innov.
– volume: 12
  start-page: 307
  year: 2019
  end-page: 392
  ident: b45
  article-title: An introduction to variational autoencoders
  publication-title: Found. Trends® Mach. Learn.
– volume: 28
  year: 2020
  ident: b1
  article-title: Roadmap on multiscale materials modeling
  publication-title: Modelling Simul. Mater. Sci. Eng.
– volume: 51
  start-page: 5985
  year: 2020
  end-page: 5999
  ident: b30
  article-title: Overview: Computer vision and machine learning for microstructural characterization and analysis
  publication-title: Met. Mater. Trans. A
– volume: 201
  year: 2021
  ident: b22
  article-title: A microstructure-informatic strategy for Vickers hardness forecast of austenitic steels from experimental data
  publication-title: Mater. Des.
– reference: .
– volume: 30
  start-page: 191
  year: 2023
  end-page: 222
  ident: b7
  article-title: A state-of-the-art review on machine learning-based multiscale modeling, simulation, homogenization and design of materials
  publication-title: Arch. Comput. Methods Eng.
– volume: 106
  start-page: 203
  year: 2018
  end-page: 227
  ident: b48
  article-title: An integrated crystal plasticity–phase field model for spatially resolved twin nucleation, propagation, and growth in hexagonal materials
  publication-title: Int. J. Plast.
– volume: 234
  start-page: 2175
  year: 2010
  end-page: 2182
  ident: b5
  article-title: Multi-scale computational homogenization: Trends and challenges
  publication-title: J. Comput. Appl. Math.
– volume: 73
  year: 2006
  ident: b4
  article-title: From electrons to finite elements: A concurrent multiscale approach for metals
  publication-title: Phys. Rev. B
– volume: 202
  year: 2021
  ident: b34
  article-title: Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder
  publication-title: Mater. Des.
– volume: 63
  start-page: 34
  year: 2011
  end-page: 41
  ident: b19
  article-title: Microstructure informatics using higher-order statistics and efficient data-mining protocols
  publication-title: JOM
– volume: 22
  year: 2020
  ident: b3
  article-title: Solving material mechanics and multiphysics problems of metals with complex microstructures using DAMASK—The Düsseldorf advanced material simulation kit
  publication-title: Adv. Eng. Mater.
– volume: 357
  year: 2019
  ident: b10
  article-title: Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network
  publication-title: Comput. Methods Appl. Mech. Engrg.
– year: 2013
  ident: b32
  article-title: Auto-encoding variational Bayes
– volume: 8
  start-page: 13461
  year: 2018
  ident: b31
  article-title: A transfer learning approach for microstructure reconstruction and structure-property predictions
  publication-title: Sci. Rep.
– volume: 179
  start-page: 317
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b23
  article-title: Hierarchical n-point polytope functions for quantitative representation of complex heterogeneous materials and microstructural evolution
  publication-title: Acta Mater.
  doi: 10.1016/j.actamat.2019.08.045
– volume: 473
  issue: 2206
  year: 2017
  ident: 10.1016/j.commatsci.2025.114145_b39
  article-title: On three-dimensional misorientation spaces
  publication-title: Proc. R. Soc. A: Math. Phys. Eng. Sci.
  doi: 10.1098/rspa.2017.0274
– volume: 8
  start-page: 190
  issue: 1
  year: 2022
  ident: 10.1016/j.commatsci.2025.114145_b37
  article-title: Learning two-phase microstructure evolution using neural operators and autoencoder architectures
  publication-title: Npj Comput. Mater.
  doi: 10.1038/s41524-022-00876-7
– volume: 8
  start-page: 13461
  issue: 1
  year: 2018
  ident: 10.1016/j.commatsci.2025.114145_b31
  article-title: A transfer learning approach for microstructure reconstruction and structure-property predictions
  publication-title: Sci. Rep.
  doi: 10.1038/s41598-018-31571-7
– volume: 218
  year: 2023
  ident: 10.1016/j.commatsci.2025.114145_b26
  article-title: Digital fingerprinting of microstructures
  publication-title: Comput. Mater. Sci.
  doi: 10.1016/j.commatsci.2022.111985
– ident: 10.1016/j.commatsci.2025.114145_b53
– volume: 202
  year: 2021
  ident: 10.1016/j.commatsci.2025.114145_b34
  article-title: Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder
  publication-title: Mater. Des.
  doi: 10.1016/j.matdes.2021.109544
– volume: 56
  start-page: 942
  issue: 5
  year: 2008
  ident: 10.1016/j.commatsci.2025.114145_b21
  article-title: Microstructure reconstructions from 2-point statistics using phase-recovery algorithms
  publication-title: Acta Mater.
  doi: 10.1016/j.actamat.2007.10.044
– volume: 150
  start-page: 212
  year: 2018
  ident: 10.1016/j.commatsci.2025.114145_b33
  article-title: Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model
  publication-title: Comput. Mater. Sci.
  doi: 10.1016/j.commatsci.2018.03.074
– volume: 17
  start-page: 395
  issue: 4
  year: 2007
  ident: 10.1016/j.commatsci.2025.114145_b51
  article-title: A tutorial on spectral clustering
  publication-title: Stat. Comput.
  doi: 10.1007/s11222-007-9033-z
– volume: 158
  start-page: 420
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b40
  article-title: DAMASK – the Düsseldorf advanced material simulation kit for modeling multi-physics crystal plasticity, thermal, and damage phenomena from the single crystal up to the component scale
  publication-title: Comput. Mater. Sci.
  doi: 10.1016/j.commatsci.2018.04.030
– volume: 35
  start-page: 155
  issue: 2
  year: 2009
  ident: 10.1016/j.commatsci.2025.114145_b49
  article-title: Metrics for 3D rotations: Comparison and analysis
  publication-title: J. Math. Imaging Vision
  doi: 10.1007/s10851-009-0161-2
– volume: 18
  start-page: 1041
  issue: 10
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b13
  article-title: Completing the picture through correlative characterization
  publication-title: Nat. Mater.
  doi: 10.1038/s41563-019-0402-8
– volume: 28
  issue: 4
  year: 2020
  ident: 10.1016/j.commatsci.2025.114145_b1
  article-title: Roadmap on multiscale materials modeling
  publication-title: Modelling Simul. Mater. Sci. Eng.
  doi: 10.1088/1361-651X/ab7150
– volume: 51
  start-page: 5985
  issue: 12
  year: 2020
  ident: 10.1016/j.commatsci.2025.114145_b30
  article-title: Overview: Computer vision and machine learning for microstructural characterization and analysis
  publication-title: Met. Mater. Trans. A
  doi: 10.1007/s11661-020-06008-4
– ident: 10.1016/j.commatsci.2025.114145_b52
– volume: 369
  year: 2020
  ident: 10.1016/j.commatsci.2025.114145_b11
  article-title: A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2020.113234
– volume: 133
  start-page: 30
  year: 2017
  ident: 10.1016/j.commatsci.2025.114145_b25
  article-title: Exploring the microstructure manifold: Image texture representations applied to ultrahigh carbon steel microstructures
  publication-title: Acta Mater.
  doi: 10.1016/j.actamat.2017.05.014
– year: 2022
  ident: 10.1016/j.commatsci.2025.114145_b46
– volume: 10
  start-page: 7164
  issue: 105
  year: 2025
  ident: 10.1016/j.commatsci.2025.114145_b43
  article-title: A python library for pre- and post-processing of DAMASK simulations
  publication-title: J. Open Source Softw.
  doi: 10.21105/joss.07164
– volume: 2
  start-page: 115
  issue: 1
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b6
  article-title: Integrating machine learning and multiscale modeling—perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences
  publication-title: Npj Digit. Med.
  doi: 10.1038/s41746-019-0193-y
– volume: 30
  start-page: 191
  issue: 1
  year: 2023
  ident: 10.1016/j.commatsci.2025.114145_b7
  article-title: A state-of-the-art review on machine learning-based multiscale modeling, simulation, homogenization and design of materials
  publication-title: Arch. Comput. Methods Eng.
  doi: 10.1007/s11831-022-09795-8
– volume: 178
  year: 2023
  ident: 10.1016/j.commatsci.2025.114145_b9
  article-title: Learning macroscopic internal variables and history dependence from microscopic models
  publication-title: J. Mech. Phys. Solids
  doi: 10.1016/j.jmps.2023.105329
– volume: 201
  year: 2021
  ident: 10.1016/j.commatsci.2025.114145_b22
  article-title: A microstructure-informatic strategy for Vickers hardness forecast of austenitic steels from experimental data
  publication-title: Mater. Des.
  doi: 10.1016/j.matdes.2021.109497
– volume: 3
  start-page: 56
  issue: 1
  year: 2014
  ident: 10.1016/j.commatsci.2025.114145_b41
  article-title: DREAM.3D: A digital representation environment for the analysis of microstructure in 3D
  publication-title: Integr. Mater. Manuf. Innov.
  doi: 10.1186/2193-9772-3-5
– volume: 73
  issue: 2
  year: 2006
  ident: 10.1016/j.commatsci.2025.114145_b4
  article-title: From electrons to finite elements: A concurrent multiscale approach for metals
  publication-title: Phys. Rev. B
  doi: 10.1103/PhysRevB.73.024108
– volume: 263
  year: 2024
  ident: 10.1016/j.commatsci.2025.114145_b38
  article-title: Trade-offs in the latent representation of microstructure evolution
  publication-title: Acta Mater.
  doi: 10.1016/j.actamat.2023.119514
– volume: 1
  start-page: 19
  year: 2002
  ident: 10.1016/j.commatsci.2025.114145_b27
  article-title: The application of principal component analysis to materials science data
  publication-title: Data Sci. J.
  doi: 10.2481/dsj.1.19
– volume: 110
  start-page: 126
  year: 2015
  ident: 10.1016/j.commatsci.2025.114145_b24
  article-title: A computer vision approach for automated analysis and classification of microstructural image data
  publication-title: Comput. Mater. Sci.
  doi: 10.1016/j.commatsci.2015.08.011
– volume: 158
  year: 2022
  ident: 10.1016/j.commatsci.2025.114145_b8
  article-title: From CP-FFT to CP-RNN: Recurrent neural network surrogate model of crystal plasticity
  publication-title: Int. J. Plast.
  doi: 10.1016/j.ijplas.2022.103430
– volume: 106
  start-page: 203
  year: 2018
  ident: 10.1016/j.commatsci.2025.114145_b48
  article-title: An integrated crystal plasticity–phase field model for spatially resolved twin nucleation, propagation, and growth in hexagonal materials
  publication-title: Int. J. Plast.
  doi: 10.1016/j.ijplas.2018.03.009
– year: 2017
  ident: 10.1016/j.commatsci.2025.114145_b44
– volume: 22
  issue: 3
  year: 2020
  ident: 10.1016/j.commatsci.2025.114145_b3
  article-title: Solving material mechanics and multiphysics problems of metals with complex microstructures using DAMASK—The Düsseldorf advanced material simulation kit
  publication-title: Adv. Eng. Mater.
  doi: 10.1002/adem.201901044
– volume: 4
  issue: 5
  year: 2016
  ident: 10.1016/j.commatsci.2025.114145_b14
  article-title: Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science
  publication-title: APL Mater.
  doi: 10.1063/1.4946894
– volume: 234
  start-page: 2175
  issue: 7
  year: 2010
  ident: 10.1016/j.commatsci.2025.114145_b5
  article-title: Multi-scale computational homogenization: Trends and challenges
  publication-title: J. Comput. Appl. Math.
  doi: 10.1016/j.cam.2009.08.077
– volume: 95
  start-page: 1
  year: 2018
  ident: 10.1016/j.commatsci.2025.114145_b16
  article-title: Computational microstructure characterization and reconstruction: Review of the state-of-the-art techniques
  publication-title: Prog. Mater. Sci.
  doi: 10.1016/j.pmatsci.2018.01.005
– volume: 63
  start-page: 34
  issue: 4
  year: 2011
  ident: 10.1016/j.commatsci.2025.114145_b19
  article-title: Microstructure informatics using higher-order statistics and efficient data-mining protocols
  publication-title: JOM
  doi: 10.1007/s11837-011-0057-7
– volume: 32
  start-page: 77
  issue: 1
  year: 2002
  ident: 10.1016/j.commatsci.2025.114145_b18
  article-title: Statistical description of microstructures
  publication-title: Annu. Rev. Mater. Res.
  doi: 10.1146/annurev.matsci.32.110101.155324
– volume: 41
  start-page: 587
  issue: 08
  year: 2016
  ident: 10.1016/j.commatsci.2025.114145_b15
  article-title: The need for microstructure informatics in process–structure–property relations
  publication-title: MRS Bull.
  doi: 10.1557/mrs.2016.163
– volume: 2
  start-page: 2
  year: 2023
  ident: 10.1016/j.commatsci.2025.114145_b47
  article-title: A novel integrated framework for reproducible formability predictions using virtual materials testing
  publication-title: Mater. Open Res.
  doi: 10.12688/materialsopenres.17516.1
– volume: 357
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b10
  article-title: Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2019.112594
– volume: 646
  start-page: 101
  year: 2015
  ident: 10.1016/j.commatsci.2025.114145_b17
  article-title: Grain-size dependent mechanical behavior of nanocrystalline metals
  publication-title: Mater. Sci. Eng.: A
  doi: 10.1016/j.msea.2015.07.075
– year: 2015
  ident: 10.1016/j.commatsci.2025.114145_b35
– volume: 8
  start-page: 93
  issue: 1
  year: 2022
  ident: 10.1016/j.commatsci.2025.114145_b2
  article-title: Modeling and simulation of microstructure in metallic systems based on multi-physics approaches
  publication-title: Npj Comput. Mater.
  doi: 10.1038/s41524-022-00764-0
– year: 2012
  ident: 10.1016/j.commatsci.2025.114145_b29
  article-title: ImageNet classification with deep convolutional neural networks
– volume: 36
  start-page: 193
  issue: 4
  year: 1980
  ident: 10.1016/j.commatsci.2025.114145_b28
  article-title: Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position
  publication-title: Biol. Cybernet.
  doi: 10.1007/BF00344251
– volume: 58
  start-page: 4432
  issue: 13
  year: 2010
  ident: 10.1016/j.commatsci.2025.114145_b20
  article-title: Optimized structure based representative volume element sets reflecting the ensemble-averaged 2-point statistics
  publication-title: Acta Mater.
  doi: 10.1016/j.actamat.2010.04.041
– start-page: 8024
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b42
  article-title: PyTorch: An imperative style, high-performance deep learning library
– volume: 12
  start-page: 307
  issue: 4
  year: 2019
  ident: 10.1016/j.commatsci.2025.114145_b45
  article-title: An introduction to variational autoencoders
  publication-title: Found. Trends® Mach. Learn.
  doi: 10.1561/2200000056
– year: 2024
  ident: 10.1016/j.commatsci.2025.114145_b36
– year: 2024
  ident: 10.1016/j.commatsci.2025.114145_b12
– volume: 2
  start-page: 433
  issue: 4
  year: 2010
  ident: 10.1016/j.commatsci.2025.114145_b50
  article-title: Principal component analysis: Principal component analysis
  publication-title: Wiley Interdiscip. Rev.: Comput. Stat.
  doi: 10.1002/wics.101
– year: 2013
  ident: 10.1016/j.commatsci.2025.114145_b32
SSID ssj0016982
Score 2.4496005
Snippet Microstructure quantification is an important step towards establishing structure–property relationships in materials. Machine learning-based image processing...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 114145
SubjectTerms Crystal plasticity
Feature learning
Fingerprinting
Machine learning
Microstructure characterisation
Surrogate model
Title 3D variational autoencoder for fingerprinting microstructure volume elements
URI https://dx.doi.org/10.1016/j.commatsci.2025.114145
Volume 259
WOSCitedRecordID wos001547379800002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 0927-0256
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0016982
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9wwEBbNpofkUNo8aNIHOvRmHCzZsqzelm5KG0pYSAJ7M5Itk0fXG7Kbx8_vjCU73m4gLaUXY8tYtmY-xqPR6BtCPglmrYRfVxjFWoVJmmShlnCZxlylldQ8FbopNiGPj7PJRI19WtG8KScg6zp7eFDX_1XV0AbKxq2zf6HurlNogHNQOhxB7XD8I8XHo-AOJsBtkE_fLmZIVomcEZhSWDVxPAznNQnPU0zIcySyuJTgjFVgXVL5vO-6uvoPbbfg6LrBBP4f-mjdfcU9n5AfjA46UC2uur1mq3fHP2f3Phw7LPU0OOrunJxrLOegm4Wm8Q36rJ5i2IcruOjysXwMbWUfjQtGchmi99W3y9xRha_YeBduuEQVwVhhkAf4HiQ9Zo6Z8jcC7RPsHTsHbw_tVbJG1rkUKhuQ9eH3w8lRt-qUqqa4WPc1S_mAT77uaW-m56Gcviav_NSCDh0k3pAXtt4imz3CyW3yIx7RHjhoDxwUwEGXwUGXwUGdxGgLjh1y9vXw9Mu30NfTCIuY8UWYlIKZOC6ZkaURklUwgjKrpI1sBbOAqDSKc8OUlcYkJgNJpbYwlbA6EloUPN4lg3pW27eEsoKVzPJEi9QkrNAKHW-ORBPQJiOxR6JWLvm1o03J23zCy7wTZY6izJ0o98jnVn65R67z6nJQ_HMP7__Lw-_IxiNS35MBiNV-IC-Lu8XF_OajB8kv04OHCQ
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=3D+variational+autoencoder+for+fingerprinting+microstructure+volume+elements&rft.jtitle=Computational+materials+science&rft.au=White%2C+Michael+D.&rft.au=Atkinson%2C+Michael+D.&rft.au=Plowman%2C+Adam+J.&rft.au=Shanthraj%2C+Pratheek&rft.date=2025-09-01&rft.pub=Elsevier+B.V&rft.issn=0927-0256&rft.volume=259&rft_id=info:doi/10.1016%2Fj.commatsci.2025.114145&rft.externalDocID=S0927025625004884
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0927-0256&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0927-0256&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0927-0256&client=summon