Ring artifact reduction in synchrotron X-ray tomography through helical acquisition

In synchrotron X-ray tomography, systematic defects in certain detector
 elements can result in arc-shaped artifacts in the final reconstructed image of the
 scanned sample. These ring artifacts are commonly found in many applications of
 synchrotron tomography, and can make it difficult or impossible to use the reconstructed
 image in further analyses. The severity of ring artifacts is often reduced in practice by
 applying pre-processing on the acquired data, or post-processing on the reconstructed
 image. However, such additional processing steps can introduce additional artifacts as
 well, and rely on specific choices of hyperparameter values.
 
 In this paper, a different approach to reducing the severity of ring artifacts is
 introduced: a helical acquisition mode. By moving the sample parallel to the rotation
 axis during the experiment, the sample is detected at different detector positions in each
 projection, reducing the effect of systematic errors in detector elements. Alternatively,
 helical acquisition can be viewed as a way to transform ring artifacts to helix-like
 artifacts in the reconstructed volume, reducing their severity. We show that data
 acquired with the proposed mode can be transformed to data acquired with a virtual
 circular trajectory, enabling further processing of the data with existing software
 packages for circular data. Results for both simulated data and experimental data show
 that the proposed method is able to significantly reduce ring artifacts in practice, even
 compared with popular existing methods, without introducing additional artifacts.

• View at Publisher