Unsupervised deep denoising for fourdimensional scanning transmission electron microscopy  

作　　者：Alireza Sadri Timothy C.Petersen Emmanuel W.C.Terzoudis-Lumsden Bryan D.Esser Joanne Etheridge Scott D.Findlay 

机构地区：[1]School of Physics and Astronomy,Monash University,Melbourne,VIC,Australia [2]Monash Centre for Electron Microscopy,Monash University,Melbourne,VIC,Australia

出　　处：《npj Computational Materials》2024年第1期660-672,共13页计算材料学(英文)

基　　金：supported under the Discovery Projects funding scheme of the Australian Research Council(Project No.FT190100619);funded by Australian Research Council grant LE170100118 and LE0454166;supported by an Australian Government Research Training Programme Scholarship.

摘　　要：By simultaneously achieving high spatial and angular sampling resolution,four dimensional scanning transmission electron microscopy(4DSTEM)is enabling analysis techniques that provide great insight into the atomic structure of materials.Applying these techniques to scientifically and technologically significant beam-sensitive materials remains challenging because the low doses needed to minimise beam damage lead to noisy data.We demonstrate an unsupervised deep learning model that leverages the continuity and coupling between the probe position and the electron scattering distribution to denoise 4D STEM data.By restricting the network complexity it can learn the geometric flowpresent but not the noise.Through experimental and simulated case studies,we demonstrate that denoising as a preprocessing step enables 4D STEM analysis techniques to succeed at lower doses,broadening the range of materials that can be studied using these powerful structure characterization techniques.

关 键 词：doses enable restrict 

分 类 号：TP1[自动化与计算机技术—控制理论与控制工程]