Constraining by a family of strictly nonexpansive idempotent functions with applications in image reconstruction (Q2376863)

The authors present the constrained version of the general iterative method which includes the algorithm of Kaczmarz, Cimmino, Jacobi projective and diagonal weighting. As an example, the authors use it to study a particular image reconstruction inverse problem arised from 3D tomographic particle image velocimetry. The idea is to seed the turbulent fluids with particles that follow the flow dynamics in a volume, capture multiple two-dimensional projections and reconstruct the 3D image. The algorithm proposed by the authors can determine in every experiment the correct number of particles and, although there are various displaced points, the obtained image are similar to the original ones.