Sinogram interpretability based CT artefact reduction for multi-material workpieces

Abstract

Computed tomography (CT) is a technology for the simultaneous inspection of internal and external features of workpieces. However, beam hardening and scattered radiation artefacts caused by the different material densities and atomic numbers commonly confound CT measurements of multi-material workpieces. Multispectral CT (MSP CT) measurements reduce artefacts by combining multiple CT volumes performed with different x-ray spectra. However, all known simulation-based CT parameter optimisation approaches for industrial applications consider only the single-spectrum case. This paper investigates the CT radon space, focusing on sinogram interpretability. Multispectral CT x-ray spectra can be determined by defining and analysing sinogram areas corresponding to equivalent volume slice artefacts. Validation indicates that the two MSP CT spectra recommended via a proposed CT simulation considering critical sinogram areas cause an artefact reduction. Results are quantified by a standard deviation reduction of measured geometrical features.