Context : X-ray micro tomography (micro-CT) is rapidly establishing itself as a standard technique for microstructural analysis in a wide range of fields, from biology to materials science. In conjunction with monochromatic X-rays acquired from a synchrotron source (SR-µCT), this yields a quantitative, fully 3D imaging technique with effective resolutions from ~20 µm down to ~100 nm. In micro-CT, a tomographic image is reconstructed from a set of its projection (Radon Transform). When the resolution is pushed higher and higher, the field of view (FOV) is diminished, then the projections are often truncated. In this case, the reconstruction is no more exact and the quantitative aspect of SR-µCT is lost. To circumvent this problem, we consider “zoom-in tomography”, where a low resolution scan laterally covering the whole object is recorded and combined with a high resolution scan in the region of interest. Due to recent development at beamline ID19 at the European Synchrotron Radiation Facility (ESRF), Grenoble, this modality has been made available on a routine basis.
Aim : The aim of this project is to implement and evaluate algorithms for zoom-in tomography. The evaluation will be performed on both simulated data and experimental data acquired at ID19, ESRF. Evaluation criteria e.g., based on resolution and signal-to-noise ratio has to be defined.
Methodology : A platform for simulation of parallel and cone beam monochromatic tomography is already in place and can be used for the simulation part of the study. Data sets of constructed objects with known composition have also been recorded at ID19, ESRF and will form the basis for the experimental study. Different tomographic reconstruction methods based will be compared. The implemented methods should further be applied for mineralization quantification of series of data acquired on bone biopsies. Requirements: Knowledge in image processing and programming language is required (Matlab, C, C++ ). The applicant should also have the capacity to integrate a multidisciplinary environment (imaging, biomedical applications, physics).