3D Image Reconstruction for Implosion Pellet in ICF Experiment Based on Iterative Algorithms

The projection images of implosion-pellets, captured by framing cameras or pinhole cameras, are two-dimensional in the inertia confinement fusion experiments. Since the two-dimensional images lack depth information, they are hardly used to diagnose the compression symmetry of an implosion-pellet. A 3D image of the implosion-pellet reconstructed from the two-dimensional projection images can overcome these problems. As the iterative algorithms can reconstruct the original 3D image from just a few projections with good noise suppression, three iterative algorithms which are commonly applied in the CT image reconstruction field are utilized to reconstruct a 3D image of an implosion-pellet. The numerical simulations show that the algebraic reconstruction technique algorithm performs best under the condition that the projection images are 'incomplete' and noise free, or with not so heavy noise. When there is heavy noise in the projection images, the simultaneous iterative reconstruction technique still surpasses the other algorithms.