These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
Machine Learning-Based Multidomain Processing for Texture-Based Image Segmentation and Analysis
preprintsubmitted on 08.01.2020, 03:08 and posted on 09.01.2020, 11:45 by Nikolay Borodinov, Wan-Yu Tsai, Vladimir V. Korolkov, Nina Balke, Sergei Kalinin, Olga S. Ovchinnikova
Atomic and molecular resolved atomic force microscopy (AFM) images offer unique insights into materials properties such as local ordering, molecular orientation and topological defects, which can be used to pinpoint physical and chemical interactions occurring at the surface. Utilizing machine learning for extracting underlying physical parameters increases the throughput of AFM data processing and eliminates inconsistencies intrinsic to manual image analysis thus enabling the creation of reliable frameworks for qualitative and quantitative evaluation of experimental data. Here, we present a robust and scalable approach to the segmentation of AFM images based on flexible pre-selected classification criteria. Usage of supervised learning and feature extraction allows to retain the consideration of specific problem-dependent features (such as types of periodical structure observed in the images and the associated numerical parameters: spacing, orientation, etc.). We highlight the applicability of this approach for segmentation of molecular resolved AFM images based on crystal orientation of observed domains, automated selection of boundaries and collection of relevant statistics. Overall, we outline a general strategy for machine learning-enabled analysis of nanoscale systems exhibiting periodic order that could be applied to any analytical imaging technique.
Read the published paper
in Applied Physics Letters