A variational inequality for total variation functional with constraint (Q5947222)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: A variational inequality for total variation functional with constraint |
scientific article; zbMATH DE number 1660653
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A variational inequality for total variation functional with constraint |
scientific article; zbMATH DE number 1660653 |
Statements
A variational inequality for total variation functional with constraint (English)
0 references
14 January 2003
0 references
variational inequality
0 references
total variation
0 references
structure of solutions
0 references
local minimizers
0 references
structure of minimizers
0 references
nonlinear inclusion
0 references
Euler-Lagrange inclusion of a functional
0 references
subdifferential
0 references
0 references
0.92138076
0 references
0.9150367
0 references
0 references
0.9056571
0 references
0.90292865
0 references
0 references
0 references
The authors study a nonlinear inclusion of the form NEWLINE\[NEWLINE\kappa \partial V(w)\ni w + \theta_0\quad \text{in}\quad L^2(0,L),\tag{1}NEWLINE\]NEWLINE where \(L\) is a positive and finite number, \(\theta_0\) is a given constant, \(\kappa\) is a (small) positive constant and \(V(w)\) is the total variation of any function \(w\) with \(|w|\leq 1\) on \([0,L]\). As easily seen, (1) is the Euler-Lagrange inclusion of the functional NEWLINE\[NEWLINEF_{\theta_0}(z):=\kappa V(z)-\frac{1}{2}\int_0^Lz^2 dx-\int_0^L\theta_0 z dx\quad\text{on}\quad L^2(0,L).NEWLINE\]NEWLINE The structure of the solution set of (1) for each prescribed constant \(\theta_0\) is studied and the expression for each solution \(w\) is given. It is shown that any solution \(w\) of (1) is piecewise constant and has a finite number of discontinuities in \([0,L]\). As an application, a more detailed characterization of the local minimizers of \(F_{\theta_0}\) in the one-dimensional case is obtained.
0 references