Kati Niinimäki

Larisa Beilina, Michel Cristofol, Kati Niinimäki

We consider the inverse problem of the simultaneous reconstruction of the dielectric permittivity and magnetic permeability functions of the Maxwell's system in 3D with limited boundary observations of the electric field. The theoretical stability for the problem is provided by the Carleman estimates. For the numerical computations the problem is formulated as an optimization problem and hybrid finite element/difference method is used to solve the parameter identification problem.

Larisa Beilina, Kati Niinimäki

We consider an inverse problem of reconstructing the conductivity function in a hyperbolic equation using single space-time domain noisy observations of the solution on the backscattering boundary of the computational domain. We formulate our inverse problem as an optimization problem and use Lagrangian approach to minimize the corresponding Tikhonov functional. We present a theorem of a local strong convexity of our functional and derive error estimates between computed and regularized as well as exact solutions of this functional, correspondingly. In numerical simulations we apply domain decomposition finite element-finite difference method for minimization of the Lagrangian. Our computational study shows efficiency of the proposed method in the reconstruction of the conductivity function in three dimensions.

Kati Niinimäki, Matti Lassas, Keijo Hämäläinen et al.

A computational method is introduced for choosing the regularization parameter for total variation (TV) regularization. The approach is based on computing reconstructions at a few different resolutions and various values of regularization parameter. The chosen parameter is the smallest one resulting in approximately discretization-invariant TV norms of the reconstructions. The method is tested with X-ray tomography data measured from a walnut and compared to the S-curve method. The proposed method seems to automatically adapt to the desired resolution and noise level, and it yields useful results in the tests. The results are comparable to those of the S-curve method; however, the S-curve method needs a priori information about the sparsity of the unknown, while the proposed method does not need any a priori information (apart from the choice of a desired resolution). Mathematical analysis is presented for (partial) understanding of the properties of the proposed parameter choice method. It is rigorously proven that the TV norms of the reconstructions converge with any choice of regularization parameter.