Variational Convolutional Autoencoders for Anomaly Detection in Scanning Transmission Electron Microscopy

Identifying point defects and other structural anomalies using scanning transmission electron microscopy (STEM) is important to understand a material's properties caused by the disruption of the regular pattern of crystal lattice. Due to improvements in instrumentation stability and electron op...

Full description

Saved in:
Bibliographic Details
Published in:Small (Weinheim an der Bergstrasse, Germany) Vol. 19; no. 16; pp. e2205977 - n/a
Main Authors: Prifti, Enea, Buban, James P., Thind, Arashdeep Singh, Klie, Robert F.
Format: Journal Article
Language:English
Published: Germany Wiley Subscription Services, Inc 01.04.2023
Subjects:
Anomalies
anomaly detection
Columns (structural)
Computer vision
convolutional neural networks
Crystal defects
Crystal lattices
electron microscopy
Electron optics
Electrons
Field of view
Image analysis
Machine learning
Nanotechnology
Point defects
Scanning electron microscopy
Scanning transmission electron microscopy
Transmission electron microscopy
variational convolutional autoencoders
electron microscopy
anomaly detection
variational convolutional autoencoders
convolutional neural networks
machine learning
ISSN:1613-6810, 1613-6829, 1613-6829
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Identifying point defects and other structural anomalies using scanning transmission electron microscopy (STEM) is important to understand a material's properties caused by the disruption of the regular pattern of crystal lattice. Due to improvements in instrumentation stability and electron optics, atomic‐resolution images with a field of view of several hundred nanometers can now be routinely acquired at 1–10 Hz frame rates and such data, which often contain thousands of atomic columns, need to be analyzed. To date, image analysis is performed largely manually, but recent developments in computer vision (CV) and machine learning (ML) now enable automated analysis of atomic structures and associated defects. Here, the authors report on how a Convolutional Variational Autoencoder (CVAE) can be utilized to detect structural anomalies in atomic‐resolution STEM images. Specifically, the training set is limited to perfect crystal images , and the performance of a CVAE in differentiating between single‐crystal bulk data or point defects is demonstrated. It is found that the CVAE can reproduce the perfect crystal data but not the defect input data. The disagreesments between the CVAE‐predicted data for defects allows for a clear and automatic distinction and differentiation of several point defect types. Automated analysis of atomic‐resolution scanning transmission electron microscopy images is demonstrated using a neural network approach. A Convolutional Variational Autoencoder (CVAE) architecture is used to detect crystal‐structure anomalies, such as point defects, in the bulk of SrTiO3 [001] with a training set that is limited to perfect bulk images. The performance of the CVAE in differentiating between bulk from point defects is quantified.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1613-6810
1613-6829
1613-6829
DOI:10.1002/smll.202205977