A computer simulation platform for the estimation of measurement uncertainties in dimensional X-ray computed tomography

The knowledge of measurement uncertainty is of great importance in conformance testing in production. The tolerance limit for production must be reduced by the amounts of measurement uncertainty to ensure that the parts are in fact within the tolerance. Over the last 5 years, industrial X-ray computed tomography (CT) has become an important technology for dimensional quality control. In this paper a computer simulation platform is presented which is able to investigate error sources in dimensional CT measurements. The typical workflow in industrial CT metrology is described and methods for estimating measurement uncertainties are briefly discussed. As we will show, the developed virtual CT (VCT) simulator can be adapted to various scanner systems, providing realistic CT data. Using the Monte Carlo method (MCM), measurement uncertainties for a given measuring task can be estimated, taking into account the main error sources for the measurement. This method has the potential to deal with all kinds of systematic and random errors that influence a dimensional CT measurement. A case study demonstrates the practical application of the VCT simulator using numerically generated CT data and statistical evaluation methods.

► We give an overview to dimensional X-ray computed tomography metrology.
► Currently used methods for uncertainty estimation of CT measurements are discussed.
► We simulate an X-ray computed tomography system for image formation and error source analysis.
► Using the Monte Carlo method enables simulation-based measurement uncertainty estimation.
► A case study using a simple workpiece demonstrates our method.