Collection Structured Regularization Summer School - 19-22/06/2017
Organizer(s)
Date(s)
25/04/2024
Towards realistic compressed sensing
By Claire Boyer
First, we will theoretically justify the applicability of compressed sensing (CS) in real-life applications. To do so, CS theorems compatible with physical acquisition constraints will be presented. These new results do not only encompass structure in the acquisition but also structured sparsity of the signal of interest. They also provide insight on the choice of the optimal sampling strategy. Then, we will present a way to generate constrained subsampling schemes that can be implemented on real sensors while mimicking sampling strategies known to be theoretically optimal but unusable in practice. This sampling scheme generation give good reconstruction results in simulation. This work relies on measure projection and will be illustrated in the case of MRI.
