Modelling an indirect imaging X-ray system to improve analysis of X-ray projection images

Submission note: A thesis submitted in total fulfilment of the requirements for the degree of Doctor of Philosophy to the School of Molecular Sciences, College of Science, Health and Engineering, La Trobe University, Bundoora.

High quality X-ray projection images are necessary to investigate the mechanisms of intergranular corrosion in Aluminium alloys that are commonly used in the field of aeronautics. The quality of the measured projection image depends on the elements of the X-ray imaging system, including the geometry, source and the detector. In this thesis, we aim to investigate the quality in terms of spatial resolution and visibility of the sample features. To do this, we present a model of an indirect X-ray imaging system that includes the source, diffraction effects, experimental geometry, model samples, a detailed representation of the imaging detector, and noise effects. We validated the model of the detector with measurements obtained using synchrotron radiation and a polychromatic source. The model of the indirect imaging system was also validated using a polychromatic source for a test pattern and a model sample that hardens the X-ray beam. The validated model was used to predict the loss of visibility of sample features for different imaging setups, source spectra and signal-tonoise ratios. It was also used to obtain an accurate point spread function (PSF) for the imaging system and a given X-ray source spectra that includes all X-ray energy dependent processes in the detector and beam hardening due to the sample. Deconvolution of the PSF from the simulated and measured projection image demonstrated significant improvement in the visibility of the features. Finally, an improvement in the quality of the measured projection image of a real-world sample shows how deconvolving the model PSF of the imaging system can be used to enhance the study of intergranular corrosion of Al alloys.