PtychoShelves, a versatile high‐level framework for high‐performance analysis of ptychographic data. Issue 2 (13th March 2020)

• Record Type: Journal Article
• Authors: Wakonig, Klaus; Stadler, Hans-Christian; Odstrčil, Michal; Tsai, Esther H. R.; Diaz, Ana; Holler, Mirko; Usov, Ivan; Raabe, Jörg; Menzel, Andreas; Guizar-Sicairos, Manuel
• Is Part Of: Journal of applied crystallography. Volume 53:Issue 2(2020)
• Page Start: 574
• Page End: 586
• Publication Date: 2020-03-13
• Subjects: ptychography -- PtychoShelves -- MATLAB -- high‐performance computing -- phase retrieval -- synchrotrons; Crystallography -- Periodicals; 548.05
• DOI: 10.1107/S1600576720001776
• Languages: English
• ISSNs: 0021-8898

Abstract

Abstract : A new computer program for analysing ptychographic data combines both high‐level simplicity and high‐performance computing on large‐scale computing clusters. It is available with a royalty‐free non‐exclusive licence for academic and non‐commercial purposes. Abstract : Over the past decade, ptychography has been proven to be a robust tool for non‐destructive high‐resolution quantitative electron, X‐ray and optical microscopy. It allows for quantitative reconstruction of the specimen's transmissivity, as well as recovery of the illuminating wavefront. Additionally, various algorithms have been developed to account for systematic errors and improved convergence. With fast ptychographic microscopes and more advanced algorithms, both the complexity of the reconstruction task and the data volume increase significantly. PtychoShelves is a software package which combines high‐level modularity for easy and fast changes to the data‐processing pipeline, and high‐performance computing on CPUs and GPUs.