.png){kind=icon}
Return
3D variational autoencoder for fingerprinting microstructure volume elements
.png){kind=icon}
OA
AI
DOI：10.1016/j.commatsci.2025.114145.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
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.
Keywords:
Machine learning
Microstructure characterisation
Feature learning
Fingerprinting
Crystal plasticity
Surrogate model
Journal
IF:
3.3
Papers:
1.4W
Citations:
3.6W
Organization
No organization information available
Cited Papers
ShareSaveShareSaveShareSaveMicrostructure informatics using higher-order statistics and efficient data-mining protocols.png)
JOM
JOMIF0
PREAI
ShareSaveShareSaveFrom electrons to finite elements: A concurrent multiscale approach for metalsPHYSICAL REVIEW BOAAI
PHYSICAL REVIEW BIF3.7
OAAIShareSaveShareSaveShareSaveOverview: Computer Vision and Machine Learning for Microstructural Characterization and Analysis
OAAI
OAAIShareSave
View more