Localization and segmentation of atomic columns in supported nanoparticles for fast scanning transmission electron microscopy  

作　　者：Henrik Eliasson Rolf Erni 

机构地区：[1]Electron Microscopy Center,Empa-Swiss Federal Laboratories for Materials Science and Technology,Überlandstrasse 129,8600 Dübendorf,Switzerland [2]Department of Materials,ETH Zürich,CH-8093 Zürich,Switzerland

出　　处：《npj Computational Materials》2024年第1期1508-1515,共8页计算材料学(英文)

基　　金：The authors acknowledge funding from the Swiss National Science Foundation(200021_196381).

摘　　要：To accurately capture the dynamic behavior of small nanoparticles in scanning transmission electron microscopy,high-quality data and advanced data processing is needed.The fast scan rate required to observe structural dynamics inherently leads to very noisy data where machine learning tools are essential for unbiased analysis.In this study,we develop a workflow based on two U-Net architectures to automatically localize and classify atomic columns at particle-support interfaces.The model is trained on non-physical image simulations,achieves sub-pixel localization precision,high classification accuracy,and generalizes well to experimental data.We test our model on both in situ and ex situ experimental time series recorded at 5 frames per second of small Pt nanoparticles supported onCeO2(111).The processedmovies show sub-second dynamics of the nanoparticles and reveal site-specific movement patterns of individual atomic columns.

关 键 词：COLUMNS ATOMIC SUPPORTED 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]