% Here we compute the SVD of the sparse tomographic measurement matrix A
% using full matrix algorithms. This is not very efficient coding, and it
% is done only for educational purposes.
%
% The following routine must be run before this file: XRMA_matrix_comp.m.
%
% Jennifer Mueller and Samuli Siltanen, October 2012
% Choose resolution
N = 16;
% Load precomputed results at resolution N
eval(['load RadonMatrix', num2str(N), ' A measang target N P Nang']);
% Compute SVD of A using full matrix algorithms
[U,D,V] = svd(full(A));
D = sparse(D);
% Save the result to file (with filename containing the resolution N)
eval(['save XRME_SVD', num2str(N), ' U D V A measang target N P Nang']);
% View the results
XRME_SVD_plot(N)