Alignment methods for nanotomography with deep sub-pixel accuracy

This repository contains codes for alignment of projections for tomography and laminography. Additionally we include artificial, i.e. simulated, and experimental examples.

Here you will find:

1) cSAXS_matlab_tomo_shared.zip - Codes and routines, as introduced in [1], for deep sub-pixel alignment and reconstruction of tomography and laminography datasets.

2) example_data.mat - artificial tomography dataset that serves as an example for the alignment codes.

Citation and acknowledgements

If you use the codes from this repository here is where to find more information and the expected citation.

For use or further development of the tomography and laminography alignment and reconstruction codes please cite:
[1] M. Odstrcil, M. Holler, J. Holler, M. Guizar-Sicairos,"Alignment methods for nanotomography with deep sub-pixel accuracy", Opt. Express, (2019).

Code requirements:

Reconstruction scripts were tested for Matlab2018a with parallel toolkit, CUDA 9.0. NVIDIA, RHEL 7.6
GPU is required for the tomographic reconstruction.

Instalation and reconstruction:

  1) Download example datasets and code from https://doi.org/10.5281/zenodo.3539550 or a measured ptychotomography dataset from  https://doi.org/10.5281/zenodo.3539513
  2) Make sure that the downloaded datasets can be loaded by matlab
  3) Choose one of the provided datasets in the "run_simple_example.m" template
  4) Run script "run_simple_example.m". The final results will be stored in folder defined in par.output_folder

Example dataset contains following two variables:

stack_object - complex valued unaligned projection with dimensions [Npix_vertical , Npix_horizontal, number_of_angles]

  theta - vector of corresponding projection angles in degrees

This code and subroutines are part of a continuous development. There is no liability on PSI or cSAXS. License for the codes can be found in the individual scripts.