This paper concerns the reconstruction of a scalar coefficient of a second-order elliptic equation in divergence form posed on a bounded domain from internal data.This problem finds applications in multi-wave imaging,...This paper concerns the reconstruction of a scalar coefficient of a second-order elliptic equation in divergence form posed on a bounded domain from internal data.This problem finds applications in multi-wave imaging,greedy methods to approximate parameter-dependent elliptic problems,and image treatment with partial differential equations.We first show that the inverse problem for smooth coefficients can be rewritten as a linear transport equation.Assuming that the coefficient is known near the boundary,we study the well-posedness of associated transport equation as well as its numerical resolution using discontinuous Galerkin method.We propose a regularized transport equation that allow us to derive rigorous convergence rates of the numerical method in terms of the order of the polynomial approximation as well as the regularization parameter.We finally provide numerical examples for the inversion assuming a lower regularity of the coefficient,and using synthetic data.展开更多
基金ANR-17-CE40-0029 of the French National Research Agency ANR(project MultiOnde).
文摘This paper concerns the reconstruction of a scalar coefficient of a second-order elliptic equation in divergence form posed on a bounded domain from internal data.This problem finds applications in multi-wave imaging,greedy methods to approximate parameter-dependent elliptic problems,and image treatment with partial differential equations.We first show that the inverse problem for smooth coefficients can be rewritten as a linear transport equation.Assuming that the coefficient is known near the boundary,we study the well-posedness of associated transport equation as well as its numerical resolution using discontinuous Galerkin method.We propose a regularized transport equation that allow us to derive rigorous convergence rates of the numerical method in terms of the order of the polynomial approximation as well as the regularization parameter.We finally provide numerical examples for the inversion assuming a lower regularity of the coefficient,and using synthetic data.
