SurfaceScience21 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

O: Fachverband Oberflächenphysik

O 42: Poster Session III: Poster to Mini-Symposium: Machine learning applications in surface science I

O 42.4: Poster

Dienstag, 2. März 2021, 10:30–12:30, P

Analysis of grazing-incidence wide-angle X-ray scattering by neural networks — •Vladimir Starostin, Alessandro Greco, Alina Pleli, Alexander Hinderhofer, Alexander Gerlach, and Frank Schreiber — Institute of Applied Physics, University of Tübingen, Germany

Grazing-incidence wide-angle x-ray scattering (GIWAXS) is an indispensable tool for studying nanostructure surfaces and thin films. It is widely used in real-time studies of thin film growth. However, high acquisition rates of real-time experiments lead to enormous amounts of data to be analysed. For instance, a modern 2D X-ray detector has around 4.5 million of pixels and produces up to 6 Gb of data per second at the maximum frame rate of 750 Hz. In the future, these numbers will only increase and it may become unfeasible to analyze or even save unprocessed data. To address these problems, some automated tools need to be developed [1].

In this work, we present a machine learning approach that provides feature detection of GIWAXS images in an automated fashion. This simplifies the experimental data analysis and might enable on-the-fly preprocessing of GIWAXS data.

[1] Greco et al., J. Appl. Cryst., 52, 1342 (2019)