This paper presents an approach to assess the quality of the data extracted with computed tomography (CT) measuring systems to perform geometrical evaluations. The approach consists in analyzing the error features introduced by the CT measuring system during the extraction operation. The analysis of the features is performed qualitatively (using graphical analysis tools) and/or quantitatively (by means of the root-mean-square deviation parameter of the error features). The approach was used to analyze four sets of measurements performed with an industrial x-ray cone beam CT measuring system. Three test parts were used in the experiments: a high accuracy manufacturing multi-wave standard, a calibrated step cylinder and a calibrated production part. The results demonstrate the usefulness of the approach to gain knowledge on CT measuring processes and improve the quality of CT geometrical evaluations. Advantages and limitations of the approach are discussed.

The occurrence of thermal drift in industrial computed tomography (CT) systems has been reported as a significant source of error on geometrical evaluations. During CT-scans, heating inside the cabinet and varying environmental conditions may affect the position of the focal spot and distort the manipulator system, leading to relative displacement of X-ray projections and distortions in the reconstructed 3D image. This paper presents an experimental investigation on influence of the thermal effects on dimensional CT measurements. A correction method based on the manipulation of the projections was developed and evaluated. The method consists in repeating the acquisition of first projection at the end of the scan and calculating the displacement vector between these projections. The remaining projections are then corrected proportionally to this displacement. The results showed a significant reduction of the roundness deviation values measured on a precision sphere after the correction.

No abstract

No abstract