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

MM: Fachverband Metall- und Materialphysik

MM 53: Data Driven Material Science: Big Data and Workflows VI

MM 53.10: Vortrag

Donnerstag, 21. März 2024, 12:45–13:00, C 243

Deep learning-based feature detection on 2D X-ray scattering data for high throughput data analysis — •Alexander Hinderhofer, Vladimir Starostin, Constantin Voelter, Alexander Gerlach, and Frank Schreiber — Institut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, Germany

In situ real-time grazing-incidence X-ray diffraction (GIXD) is a key technique for thin film structural characterization during sample preparation. In-situ GIXD can produce large amounts of data, on the scale of several thousand images per hour, frequently exceeding the capabilities of traditional data processing methods. We propose an automated pipeline for the analysis of GIXD images, based on the Faster Region-based Convolutional Network architecture for object detection, modified to conform to the specifics of the scattering data. The model exhibits high accuracy in detecting diffraction features on noisy patterns with various experimental artifacts. We demonstrate our method on real-time tracking of organic-inorganic perovskite structure crystallization. By design, our approach is equally suitable for other crystalline thin-film materials.[1] Further, we discuss a high quality GIXD dataset with more than 1600 labeled features for performance evaluation of feature detection models in GIXD.

[1] V. Starostin et al. npj Comput Mater 8 (2022) 101

Keywords: Data analysis; Machine learning; X-ray scattering; High-throughput experiments