Welcome to odftomo’s documentation!

Introduction

Texture tomography is a way of reconstructing data from scanning synchrotron x-ray diffraction computed tomography (XRD-CT) experiments. The goal is to spatially map the microstructure of a polycrystalline sample by reconstructing an orientation distribution function in each voxel of a 3D sample.

This python package was developed at the Paul Scherrer Institut for the publication Texture tomography with high angular resolution utilizing sparsity as a set of plugins for mumott but has since grown a bit in scope.

The main workflow in texture tomography consists of three steps:

• Azimuthal Integration
  • Crystallography
  • Azimuthal bins
  • Background subtraction
  • Transmission correction
  • Scaling factors
• Reconstruction
  • Orientation distribution functions
  • Geometry parameters
  • Running the reconstruction
• Analysis
  • Scalar quanitites
  • Single voxel ODFs
  • Orientation field
  • Grain segmentation

A few specialized workflows are documented in separate jupyter notebook:

• Grain-by-grain workflow
• Centroid orientations and clustering algorithms

The code is hosted on gitlab.

Installation

The package is not on pypi, so you have to download it from git. This can eg. look like:

cd folder_where_I_want_the_python_code
git clone https://gitlab.com/liebi-group/software/odf-mumott.git
python -m pip install ./odf-mumott

The dependencies are all the dependencies of mumott: mainly numpy, scipy, matplotlib and numba. The use of numba means you might have to downgrade numpy, for which reason you might want to use a fresh virtual environment.

The notebooks have been tested with python 3.9 and 3.11.

Furthermore, to run some of the jupyter-notebooks, you will need jupyter and pyFAI.

Miscellaneous

• Conventions