Correcting for probe wandering by precession path segmentation

Scanning precession electron diffraction datasets and processing scripts used in the journal publication "Correcting for probe wandering by precession path segmentation".

DOI link to publication: https://doi.org/10.1016/j.ultramic.2023.113715

Prerequisites

To run the scripts necessary to process the data, the open source packages JupyterLab and HyperSpy need to be installed. These notebooks were created with these package versions:

hyperspy 1.6.4

jupyterlab 3.2.0

Data files

There are three data types:

Scanning precession electron diffraction (SPED) datasets contain the .hspy extension. There are two SPED datasets acquired by precession path segmentation, and one regular dataset:

SPED_256x256_22x22_10186nm_NBD_a5_spot1nm_CL20cm_125msExp_3000msFB_subframing_x8_pivotoff_01.hspy is a precession path segmentation dataset presented in figure 2 in the article.

SPED_zoom1_256x256_12x12_5556nm_NBD_a5_spot1nm_CL20cm_10msExp_3000msFB_subframing_x8.hspy is a precession path segmentation dataset presented in figures 1 and 3 in the article.

SPED_zoom1_256x256_12x12_5556nm_NBD_a5_spot1nm_CL20cm_10msExp_300msFB.hspy is a regular SPED dataset of the same region as the dataset just above (also containing "zoom1" in its name).

A series of diffraction patterns with de-rocking switched off can be found in the eight_segments.npy file. This dataset was recorded by Dr. Tina Bergh at Department of Chemical Engineering, Norwegian University of Science and Technology, and is meant to be used as an illustration of the precession segments used in figure 1 in the article.

Virtual bright field images of the precession path segmented scans, before and after rigid correction in SmartAlign, can be found with the .png extension. The files starting with "pivot01..." are images from the precession path segmentation scan with an intentional pivot point misalignment, while the files starting with "small..." are from the well aligned scan. The files ending with "_reg" are regular, i.e., non-corrected compound VBF images, while the files ending with "_cor" are SmartAlign rigidly corrected VBF images.

Processing scripts

There are two processing scripts:

p01_slicing_segments.ipynb is used to process the precession path segmentation datasets, from slicing of the raw data to constructing virtual bright field images.

p02_image_processing.ipynb is used to create the images found in the article and for image analysis. The latter includes blur quantification and measuring edge sharpness.