Analysis of the effect of cone-beam geometry and test object configuration on the measurement accuracy of a computed tomography scanner used for dimensional measurement

Jagadeesha Kumar, Alex Attridge, Paul Wood, Mark A. Williams

Measurement Science and Technology · 2011 · 64 citations · 14 references

Concepts

TL;DR

Industrial CT scanners enable non‑contact dimensional measurement of small, fragile components and hard‑to‑access internal features, but accuracy and repeatability depend on cone‑beam geometry, test‑object configuration, x‑ray power, material, size, detector characteristics, and data‑analysis methods. This study investigates how errors in system geometry and test‑object configuration affect measurement accuracy across the entire scan volume of a cone‑beam CT scanner. A cone‑beam simulation model incorporating image projection and reconstruction was used, with deliberate geometric parameter errors introduced to evaluate their impact on measurement accuracy for various positions, orientations, and sizes of the test object. The simulations reveal that geometric errors markedly affect dimensional measurements at specific test‑object configurations, underscoring the need for precise system alignment and accurate parameter estimation to achieve reliable CT measurements.

Abstract

Industrial x-ray computed tomography (CT) scanners are used for non-contact dimensional measurement of small, fragile components and difficult-to-access internal features of castings and mouldings. However, the accuracy and repeatability of measurements are influenced by factors such as cone-beam system geometry, test object configuration, x-ray power, material and size of test object, detector characteristics and data analysis methods. An attempt is made in this work to understand the measurement errors of a CT scanner over the complete scan volume, taking into account only the errors in system geometry and the object configuration within the scanner. A cone-beam simulation model is developed with the radiographic image projection and reconstruction steps. A known amount of errors in geometrical parameters were introduced in the model to understand the effect of geometry of the cone-beam CT system on measurement accuracy for different positions, orientations and sizes of the test object. Simulation analysis shows that the geometrical parameters have a significant influence on the dimensional measurement at specific configurations of the test object. Finally, the importance of system alignment and estimation of correct parameters for accurate CT measurements is outlined based on the analysis.

References

14