Invited Talk by Prof. Per Christian Hansen

Column-Action Methods in Image Reconstruction

Abstract: Column-oriented versions of algebraic iterative methods are interesting alternatives to their row-version counterparts: they converge to a least squares solution, and they provide a basis for saving computational work by skipping small updates. We motivate these methods from an optimization point of view, we present a convergence analysis, and we discuss two techniques (loping and flagging) for reducing the work. The performance of the algorithms is illustrated with numerical examples from computed tomography.

Biography: Professor Per Christian Hansen has worked with numerical regularization algorithms for 30 years, has published 100 papers in leading journals, and his books on numerical methods for inverse problems are widely used. He has developed several related software packages, of which Regularization Tools (now in its 4th version) is a popular toolbox for analysis and solution of discrete inverse problems. His h-index is 44 according to Google Scholar (September 2016) and he is a SIAM Fellow. His current research projects involve algorithms for tomographic reconstruction (joint with DTU Physics), electrical impedance tomography with hybrid data, and iterative image deblurring algorithms (joint with Emory University, Atlanta, USA, and Tufts University, Boston, USA).