Evaluation of digital models for dimensional X-ray computed tomography – analysis of a multi feature check measurement

Abstract Digital models using computer simulation for the acquisition of X-ray projections offer the possibility to virtually perform a X-ray computed tomography (CT) measurement. Virtual CT scans benefit research and development of X-ray computed tomography, serving various purposes while reducing physical scan time and therefore cost. The accuracy of a digital model determines its utility. Suitable test procedures of digital models can assure the sufficient accuracy of a digital model regarding a specific task. For the use of digital models in dimensional metrology an agreement of dimensional measurands compared to a physical counterpart is important. In this contribution, a physical and virtual CT measurements of a calibrated Multi Feature Check mini specimen are evaluated regarding surface transition metrics of the volume data and its metrological compatibility of dimensional measurands. For the evaluation of the gray value response at the surface transition, a methodology that uses local sampling patterns to generate a locally averaged edge response is presented. The digital model shows promising results to serve as a basis for further extension to determine the measurement uncertainty of dimensional CT measurements using the Monte-Carlo method in the future.