An experimental study on the influence of scatter and beam hardening in x-ray CT for dimensional metrology

Abstract

Scattered radiation and beam hardening introduce artefacts that degrade the quality of data in x-ray computed tomography (CT). It is unclear how these artefacts influence dimensional measurements evaluated from CT data. Understanding and quantifying the influence of these artefacts on dimensional measurements is required to evaluate the uncertainty of CT-based dimensional measurements. In this work the influence of scatter and beam hardening on dimensional measurements is investigated using the beam stop array scatter correction method and spectrum pre-filtration for the measurement of an object with internal and external cylindrical dimensional features. Scatter and beam hardening are found to influence dimensional measurements when evaluated using the ISO50 surface determination method. On the other hand, a gradient-based surface determination method is found to be robust to the influence of artefacts and leads to more accurate dimensional measurements than those evaluated using the ISO50 method. In addition to these observations the GUM method for evaluating standard measurement uncertainties is applied and the standard measurement uncertainty due to scatter and beam hardening is estimated.

A simulation-based study on the influence of beam hardening in X-ray computed tomography for dimensional metrology

“…It should be noted the key observations in this work have been validated in an initial experimental study [30]. A more rigorous experimental study that also considers the influence of scatter is to follow.…”

Deep Scatter Estimation (DSE): Accurate Real-Time Scatter Estimation for X-Ray CT Using a Deep Convolutional Neural Network

“…Scattered X-rays contributing to the measured signal lead to a violation of this assumption, and thus, to the introduction of CT artifacts [6,9,29]. These artifacts correspond to a degradation of image quality and impair dimensional measurements [12,13]. Especially, in case of high scatter-to-primary ratios, apropriate scatter correction is crucial to avoid a loss of accuracy of the metrological assessment.…”

Interlaboratory comparison of a physical and a virtual assembly measured by CT

“…Three participants were also able to reduce the measurement-deviations-to-voxelsizes ratio down to 0.1, representing the limit of the surface determination accuracy [2]. It is believed that the small measurement errors are a direct consequence of a small extent of imaging artefacts, as observed by looking at the material greyscale distribution according to [12]. The greyscale distributions for those participants presented negligible distortions, yielding a good contrast of inner and outer edges of the physical assembly, as shown in Figure 7a.…”

“…The greyscale distributions for those participants presented visible distortions, as shown in Figure 7b. The possible causes of such image distortions can be the X-ray scatter and beam hardening [1][2] [7] [12] and the limitation of the reconstruction algorithms occurring at high magnification levels [1] [7]. After checking the CT data sets provided by those participants, encoder offset, temperature variation, software beam hardening correction, and misalignment of X-ray tube filament were identified as influence factors.…”