Image reconstruction from a small number of projections

摘要: Image reconstruction from projections suffers an inherent difficulty: there are different images that have identical in any finite number of directions. However, by identifying the type image is likely to occur application area, one can design algorithms may be efficacious area even when small. One such approach uses total variation minimization. We report on algorithm based this approach, and show sometimes it produces medically-desirable reconstructions computerized tomography (CT) a small projections. we also demonstrate not guaranteed provide medically-relevant information: data collected actual CT scanner for projections, noise very well result tumor brain being visible reconstruction.