A simulation-based study on the influence of the x-ray spectrum on the performance of multi-material beam hardening correction algorithms

Abstract

Listen

Translate article

Beam hardening artefacts caused by the polychromatic nature of the x-ray spectra are known to deteriorate the reconstructed image quality in multi-material industrial computed tomography. A variety of beam hardening correction (BHC) algorithms exist. Most of these methods rest on the x-ray spectra to a certain extent, which means their performance may be hindered if the spectral information is not accurate. The dependence of these methods on the spectral information, however, has not been benchmarked. This work addresses the need for such investigation by applying two sets of spectra—(1) a set of the true spectra used to produce the radiography, and (2) a set of approximated spectra acquired from simulation—to three multi-material BHC algorithms of different types. The algorithms are a segmentation based linearisation algorithm, a dual-energy algorithm, and an iterative reconstruction algorithm. Our objective in this study is to estimate the dependence of these three algorithms on spectral information. For comparable accuracy, multiple metrics are employed to quantify the performance of the methods in terms of artefact presence and dimensional metrology. The results show that under the same initial conditions, dual-energy appears to be the most sensitive one to the spectral change. Contrariwise, the segmentation method is least spectrally sensitive. The iterative method is stable over the spectral change, but performs poorly in dimensional metrology.