Segmentation of Static and Dynamic Atomic-Resolution Microscopy Data Sets with Unsupervised Machine Learning Using Local Symmetry Descriptors

• Published in: Microscopy and microanalysis Vol. 27; no. 6; pp. 1454 - 1464
• Main Authors: Wang, Ning, Freysoldt, Christoph, Zhang, Siyuan, Liebscher, Christian H., Neugebauer, Jörg
• Format: Journal Article
• Language: English
• Published: New York, USA Cambridge University Press 01.12.2021; Oxford University Press
• Subjects: Algorithms; Cluster analysis; Clustering; Datasets; Feature extraction; Image processing; Image segmentation; Image transmission; Intermetallic compounds; Learning algorithms; Machine learning; Microscopy; Neural networks; Noise; Phase transitions; Scanning transmission electron microscopy; Software and Instrumentation; Symmetry; Transmission electron microscopy; Unsupervised learning; symmetry descriptors; segmentation; HAADF-STEM; unsupervised learning
• ISSN: 1431-9276; 1435-8115; 1435-8115
• DOI: 10.1017/S1431927621012770

We present an unsupervised machine learning approach for segmentation of static and dynamic atomic-resolution microscopy data sets in the form of images and video sequences. In our approach, we first extract local features via symmetry operations. Subsequent dimension reduction and clustering analysis are performed in feature space to assign pattern labels to each pixel. Furthermore, we propose the stride and upsampling scheme as well as separability analysis to speed up the segmentation process of image sequences. We apply our approach to static atomic-resolution scanning transmission electron microscopy images and video sequences. Our code is released as a python module that can be used as a standalone program or as a plugin to other microscopy packages.