WalnutReconstructionCodes

This is a collection of Python and MATLAB scripts for loading, pre-processing and reconstructing X-ray CT projection data of 42 walnuts as described in

Henri Der Sarkissian, Felix Lucka, Maureen van Eijnatten, Giulia Colacicco, Sophia Bethany Coban, Kees Joost Batenburg, "A Cone-Beam X-Ray CT Data Collection Designed for Machine Learning", Sci Data 6, 215 (2019) or arXiv:1905.04787 (2019)

Henri Der Sarkissian, Felix Lucka, Maureen van Eijnatten, Giulia Colacicco, Sophia Bethany Coban, Kees Joost Batenburg, "A Cone-Beam X-Ray CT Data Collection Designed for Machine Learning", or arXiv:1905.04787 (2019)

• FDKReconstruction.m and FDKReconstruction.py compute FDK reconstructions for data from a single source-detector orbit, which leads to high cone angle artifacts.
• GroundTruthReconstruction.m and GroundTruthReconstruction.py compute an iterative reconstructions using the data from all three source-detector orbits, which leads to a reconstruction free of high cone angle artifacts.
• The complete data set can be found via the following links: 1-8, 9-16, 17-24, 25-32, 33-37, 38-42.

Requirements

• To use the scrips, you will either need the the ASTRA toolbox more recent than v2.1 or it's development version more recent than 1.9.0dev. If you are using conda, the latter is available through the astra-toolbox/label/dev channel.
• GroundTruthReconstruction.m makes use of the SPOT toolbox.

Contributors

Henri Der Sarkissian ([email protected]), Felix Lucka ([email protected]), CWI, Amsterdam