A phase field method for tomographic reconstruction from limited data

Classical tomographic reconstruction methods fail for problems in which there is extreme temporal and spatial sparsity in the measured data. Reconstruction of coronal mass ejections (CMEs), a space weather phenomenon with potential negative effects on the Earth, is one such problem. However, the topological complexity of CMEs renders recent limited data reconstruction methods inapplicable. We propose an energy function, based on a phase field level set framework, for the joint segmentation and tomographic reconstruction of CMEs from measurements acquired by coronagraphs, a type of solar telescope. Our phase field model deals easily with complex topologies, and is more robust than classical methods when the data are very sparse. We use a fast variational algorithm that combines the finite element method with a trust region variant of Newton's method to minimize the energy. We compare the results obtained with our model to classical regularized tomography for synthetic CME-like images.