Computed tomography is a very widespread approach to visualize multi-phase materials in 3D. This project will explore novel mathematical methods that combine several modalities, in particular X-ray CT (XCT) and Neutron CT (NCT) data to improve the resulting reconstruction and subsequent classification of a sample. Currently, reconstruction methods that treat each data modality separately are being used, from which the reconstructions are then combined into a single segmentation. This does not take into account the different nature of the data. In the project, we will study variational regularization methods that integrate prior information and use statistical models for the measurement errors. We will also use a so-called material decomposition method that expands the data in terms of a basis function for each material.

In order to demonstrate the methods, the project will team up with an existing collaboration between the 3D Imaging Center at DTU, NBI and the planetary science group at GLOBE, KU. These groups study the Mars meteorite NWA 7034, “Black Beauty” with the purpose of gaining understanding about planetary formation and habitability. With the newly developed methods, the hope is that for the first time we will be able to segment the grains of this large meteorite into distinct minerals. This type of non-destructive characterization is vital, as the meteorite is one-of-a-kind. If successful, this type of analysis is relevant for characterization of the samples returned to Earth from the Mars Perseverance mission in 2030, and as such the DDSA funded science may have an impact for ESA.