Statistical Image Analysis of Tomograms with Application to Fibre Geometry Characterisation

Research output: Book/ReportPh.D. thesisResearch

276 Downloads (Pure)


The goal of this thesis is to develop statistical image analysis tools to characterise the micro-structure of complex materials used in energy technologies, with a strong focus on fibre composites. These quantification tools are based on extracting geometrical parameters defining structures from 2D and 3D images, especially acquired through X-ray computed tomography (CT). Fibre composites are extensively used in transportation and energy technologies such as wind turbines. It is of high importance to characterise composites accurately and to understand their behaviour under load to ensure efficiency and longevity of these technologies.
Imaging with X-ray CT has been the foundation of the thesis. This enables analysis in 3D and at the micro-scale, where individual fibres are distinguishable. Additionally, ultra-fast X-ray CT and in-situ loading environments are able to image these composites with high resolution both in space and time to observe fast micro-structural changes.
This thesis demonstrates that statistical image analysis combined with X-ray CT opens up numerous possibilities for understanding the behaviour of fibre composites under real life conditions. Besides enabling characterisation of material properties, estimating individual fibre centre lines and diameters allows for quantification of small micro-structural changes with a high degree of accuracy, as it is possible to follow how each individual fibre changes across data-sets acquired under progressive loading conditions. Finally, the thesis demonstrates the precision to which fibre geometry can be characterised through X-ray CT and the developed data analysis tools.
Original languageEnglish
PublisherDTU Compute
Number of pages192
Publication statusPublished - 2018
SeriesDTU Compute PHD-2017


Cite this