Sarepy’s documentation

Welcome to Sarepy’s documentation about removing ring artifacts in tomography. This documentation is not only to explain functions used in the Sarepy package but also to cover the following topics:

• Causes of ring artifacts.

• Classification of ring artifacts.

• How ring removal methods work; pros and cons of each method; how to use them efficiently; and how to improve them.

• Applications of the methods.

Source code:

• https://github.com/nghia-vo/sarepy

• https://github.com/algotom/algotom

Author: Nghia Vo, NSLS-II, Brookhaven National Lab, US; Diamond Light Source, UK.

Keywords: Ring artifact removal, Tomography, Stripe artifacts, Tutorial.

Important notice

Starting 05/2021, methods in Sarepy have been integrated and developed further in the Algotom package, https://github.com/algotom/algotom . Algotom (Ref. [3]) is a complete package for processing tomographic data. It is installable using Conda and Pip. Demonstrations of using different removal methods on challenging sinograms are shown here: https://algotom.readthedocs.io/en/latest/toc/section4/section4_4.html

Table of Contents

• 1. Causes of ring artifacts
  • 1.1. Irregular responses of a detecting system
  • 1.2. Mismatched pixels in flat-field image
• 2. Classification of stripe artifact types
  • 2.1. Full stripe
  • 2.2. Partial stripe
  • 2.3. Unresponsive stripe
  • 2.4. Fluctuating stripe
  • 2.5. Other types of stripes
• 3. Methods for removing ring artifacts
  • 3.1. Preprocessing methods
    • 3.1.1. Equalization-based methods for removing partial and full stripe artifacts
    • 3.1.2. Prior methods for removing stripe artifacts
    • 3.1.3. Method for locating stripe artifacts
    • 3.1.4. Method for removing large stripes
    • 3.1.5. Method for removing unresponsive and fluctuating stripes
    • 3.1.6. Combinations of methods
  • 3.2. Postprocessing methods
    • 3.2.1. Fft-based methods for removing ring artifacts
  • 3.3. Common side effects of ring artifact removal methods
• 4. Applications
  • 4.1. Tomographic reconstruction pipeline for testing ring removal methods
  • 4.2. Deriving an interpolation-based ring removal method
    • remove_stripe_based_interpolation()
• 5. Improvements and new features available in Algotom
  • 5.1. Improvements
  • 5.2. Tools for designing ring removal methods
    • 5.2.1. Back-and-forth sorting
    • 5.2.2. Separation of frequency components
    • 5.2.3. Polynomial fitting along an axis
    • 5.2.4. Wavelet decomposition and reconstruction
    • 5.2.5. Stripe interpolation
    • 5.2.6. Transformation between Cartesian and polar coordinate system
    • 5.2.7. Transformation between sinogram space and reconstruction space

References

1. Nghia T. Vo, Robert C. Atwood, and Michael Drakopoulos, “Superior techniques for eliminating ring artifacts in X-ray micro-tomography,” Optics Express 26, 28396-28412 (2018); https://doi.org/10.1364/OE.26.028396.

2. Nghia T. Vo, Robert C. Atwood, and Michael Drakopoulos, “Preprocessing techniques for removing artifacts in synchrotron-based tomographic images,” Proc. SPIE 11113, Developments in X-Ray Tomography XII, 111131I (10 September 2019); https://doi.org/10.1117/12.2530324.

3. Nghia T. Vo, Robert C. Atwood, Michael Drakopoulos, and Thomas Connolley, “Data processing methods and data acquisition for samples larger than the field of view in parallel-beam tomography,” Opt. Express 29, 17849-17874 (2021); https://doi.org/10.1364/OE.418448.

License

Apache License, Version 2.0.