Nonlocal Approximation of Nonisotropic Free-Discontinuity Problems

From MaRDI portal





A large class of problems in fracture mechanics, image segmentation, liquid cristals theory can be formulated as ``free-discontinuity problems in which one looks for the minimum of integral functionals of the form: \[ \int_{\Omega}g(x,\nabla u) dx + \int_{S_u}\varphi(x,[u],\nu_u) d{\mathcal{H}}^{n-1}, \] where \(\Omega\) is an open and bounded set in \({\mathbb R}^n\), \(u \in \text{SBV}(\Omega)\) (the class of special functions of bounded variation introduced by De Giorgi and Ambrosio) and the symbols \(\nabla u, S_u,[u]\) denote, respectively, the approximate gradient, the jump set and the jump of the function \(u\), while \(\nu_u\) is the approximate normal to \(S_u\). The authors generalize an earlier result of Braides and Dal Maso cocerning the approximation (in the sense of \(\Gamma\)-convergence) of such functionals by nonlocal functionals of special form. Namely, under some natural assumptions on functions \(f\) and \(\varphi\), their functionals take the form \[ {1 \over \varepsilon} \int_{\Omega}f({1 \over {\varepsilon ^{n-1}|C|}} \int_{{x-\varepsilon C}\cap \Omega}g(y,Du) dy) dx \] with suitable function \(f\) and bounded, open and convex set \(C\). Some penalization term can also be added.











This page was built for publication: Nonlocal Approximation of Nonisotropic Free-Discontinuity Problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4258464)