Digital X-ray imaging techniques such as radiography or computerised tomography produce images of the interior of a sample. It is difficult to empirically find equipment settings such that the obtained images arc of high quality. This work presents an optimisation procedure for finding the optimal equipment settings for the imaging task at issue. It includes a suggested procedure for finding the optimiscd equipment settings even when complex samples arc investigated. To be able to mathematically optimisc the image quality, it is necessary to have a model of the X-ray imaging system together with an appropriate measure of image quality. This work proposes the ratios geometry-sensitivity/noise, density-sensitivity/noise and mass attenuation sensitivity/noise as measures of the physical image quality. A mathematical model of the imaging system was used to calculate and predict the ratios. The model predictions agreed well with the measured values. It is shown that the geometry- and density-sensitivity/noise ratios can be related to the signal-to-noise ratio and the traditional thickness (contrast) sensitivity.
The weighted maximin method is used to obtain the optimal equipment settings. With the optimal settings, no improvement can be made without worsening at least one other sensitivity/noise ratio. It is demonstrated how the weights (penalties) can be selected to focus the sensitivity on different types of features in the investigated sample.
The optimisation procedure is demonstrated on thermal barrier coatings investigated with CT. Optimised and non-optimiscd settings were used in the investigation. By optimising the equipment settings, the sensitivity/noise ratios are increased with approximately 100% for the selected X-ray path. With the optimised settings, the obtained CT images are of better quality and more features in the microstructure can be observed.