|International Symposium on Computerized Tomography for ...|
Most industrial imaging systems are equipped with X-ray tubes that produce X-ray photons with an energy distribution, that is an energy spectrum, and partly absorbing energy integrating detectors such as scintillator screens. To emphasize detectability of a defect in the final CT image the radiological differences between the defect and base material most increase. The X-ray energy spectrum has to be chosen with tube potential and shaped with filters so that the contrast from a defect within an object increases more than the fundamental image noise in the detector used. This means that the optimal image parameter settings will change with the imaging task. Each individual defect for each individual object material and geometry has its own optimal X-ray tube potential and filter. Optimal parameter settings are, however, tedious to find empirically - particularly for industrial applications, due to the wide range of imaging tasks, that is objects geometry and material.
Within this work mathematical models of the image and data collection process for radiography and CT have been developed. The objective has been to develop a simulation environment of the radiation physics in CT and radiography that considers the poly-energetic dependence in the imaging process. This simulation environment include: treatment of full X-ray energy spectrum; how it changes on its way to the detector due to absorption and scatter in the filter and the object/defect, and also the detector energy response. The simulation environment is used for fast and cost-effective studies of how parameter variations affect final image quality and indirectly - the defect detectability. In this particular case, the simulations have been applied on a high resolution CT application to determine optimal operator parameter settings that maximize the detectability of different defect types in circular objects and to predict the minimum detectable size as a function of object diameter. The simulation environment has also been used to correct for beam hardening artifacts in CT-images.
There are several important partial results. (1) Definition of quality of the CT-data in relation to the imaging task, including a model of the X-ray paths and how it is used to predict the optimal performance. (2) A model and method to determine how the information of the imaged object transfer from the detector entrance screen through the detector chain to CT projection data and further on to the final CT-image without detailed knowledge of each stage in the detector chain. (3) A model and method of how the total un-sharpness of the CT-system is determined, in terms of modulation-transfer-function as a function of spatial frequency. Finally, (4) the commonly used contrast-detail-curve, together with the limiting perception factor for detection of small details was here developed to the more useful detectable detail - object diameter diagram. Absolute X-ray spectra, for the system to be optimized, were measured using a Compton scattering spectrometer. The detector energy response was determined using Monte Carlo photon transport simulations. It was further shown that X-ray source leak radiation increases image noise when the generated X-ray spectrum contains photons with energies over the K-edges of the enclosure wall material.