Identification of discontinuous parameters by regularization for curve representations (Q4943181)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1415932
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Identification of discontinuous parameters by regularization for curve representations |
scientific article; zbMATH DE number 1415932 |
Statements
Identification of discontinuous parameters by regularization for curve representations (English)
0 references
19 November 2000
0 references
parameter identification
0 references
curve representations
0 references
regularization
0 references
discontinuous solutions
0 references
0.8537305
0 references
0.85056204
0 references
0.81262684
0 references
0.8053223
0 references
0.79404145
0 references
0.7933219
0 references
0 references
0 references
0.78365195
0 references
0.7819831
0 references
In 1998 \textit{A. Neubauer} and \textit{O. Scherzer} [SIAM J. Appl. Math. 58, No. 6, 1891-1900 (1998; Zbl 0916.65128)] developed a new approach for regularizing problems with discontinuous solutions applied to linear deblurring and denoising problems in signal processing and called it regularization for curve representations. The main idea of this approach is to represent a discontinuous function as a curve with parametrization. This allows a combination with nonlinear Tikhonov regularization in Hilbert spaces. In the paper of Kindermann and Neubauer this method is applied to nonlinear ill-posed problems, in particular to parameter identification problems where the parameter has discontinuities. The advantage of this method compared with other approaches is that no information on the number of discontinuities and their locations is required. A convergence theory is given, especially the authors prove convergence of the finite-dimensional Tikhonov regularized solutions. The theory is applied to the so-called `c-problem' aimed at identifying a parameter function \(c\) with bounded variation from measurements \(u\) satisfying the state equation \[ -(cu_x)_x=f, \quad u(0)= u(1)=0. \] A numerical case study completes the well-written paper.
0 references