MS 16: Beyond filtered backprojection: Radon inversion with a priori knowledge
Fri, 31 March, 2017, 13:30–15:30, Room: UC 202G
Organizers
Martin Benning, Matthias J. Ehrhardt, and Carola Schönlieb
Abstract
Since the discovery of the Radon transform 100 years ago the inversion methods for the Radon transform have changed significantly. As the problem of inverting the Radon transform is ill-posed, regularisaton strategies are necessary to compute sensible approximate solutions. Classical ways of regularising the Radon inversion are based on the filtered backprojection. Subsequently, variational methods have been employed to regularise the inverison by incorporating a priori knowledge. While initially, the variational inversion still was linear, over the years it has shifted to non-linear inversion, for instance with a priori sparsity information. This minisymposium brings together researchers to present current progress on algorithms, methods and models for state-of-the-art Radon inversion.
List of speakers
Joshua Greenhalgh Solving the Polychromatic inverse problem for X-ray CT |
Kristian Bredies Direct reconstruction preconditioners for iterative variational Radon inversion |
Yiqiu Dong Directional Regularization in CT Reconstruction |
Matthias J. Ehrhardt Faster PET Reconstruction with a Stochastic Primal-Dual Hybrid Gradient Method |
Ozan Öktem Shape based prior information in image reconstruction |
Alex Sawatzky Modern Radon inversion from the perspective of industrial X-ray CT |