Data-parallel tomographic reconstruction: A comparison of filtered backprojection and direct Fourier reconstruction
J. B. T. M. Roerdink and M. A. Westenberg. Data-parallel tomographic reconstruction: A comparison of filtered backprojection and direct Fourier reconstruction. Parallel Computing, 24:2129-2142, 1998.
We consider the parallelization of two standard 2D reconstruction algorithms, filtered backprojection and direct Fourier reconstruction, using the data-parallel programming style. The algorithms are implemented on a Connection Machine CM with 16 processors and a peak performance of 2 Gflop/s. Direct Fourier reconstruction is easy to parallelize on the CM. The filtered backprojection algorithm presents a little more difficulty, due to the distributed memory of the CM. Communication is reduced by using lookup tables for the filtered projections. Reconstruction times for a 512 x 512 image are about 8.7 seconds for filtered backprojection and 1.3 seconds for direct Fourier reconstruction. Since the quality of direct Fourier reconstruction is lower than that of filtered backprojection, some extra time can be spent for improvement of the interpolation needed in the latter algorithm.
Download in gzipped postscript format© Parallel Computing. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the publisher.
This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.