We consider the Perona-Malik functional in dimension one, namely an integral functional whose Lagrangian is convex-concave with respect to the derivative, with a convexification that is identically zero. We approximate and regularize the functional by adding a term that depends on second order derivatives multiplied by a small coefficient.We investigate the asymptotic behavior of minima and minimizers as this small parameter vanishes. In particular, we show that minimizers exhibit the so-called staircasing phenomenon, namely they develop a sort of microstructure that looks like a piecewise constant function at a suitable scale.Our analysis relies on Gamma-convergence results for a rescaled functional, blow-up techniques, and a characterization of local minimizers for the limit problem. This approach can be extended to more general models.
Gobbino, M., Picenni, N. (2023). A QUANTITATIVE VARIATIONAL ANALYSIS OF THE STAIRCASING PHENOMENON FOR A SECOND ORDER REGULARIZATION OF THE PERONA-MALIK FUNCTIONAL. TRANSACTIONS OF THE AMERICAN MATHEMATICAL SOCIETY, 376(8), 5307-5375 [10.1090/tran/8841].
A QUANTITATIVE VARIATIONAL ANALYSIS OF THE STAIRCASING PHENOMENON FOR A SECOND ORDER REGULARIZATION OF THE PERONA-MALIK FUNCTIONAL
Picenni N.
2023-01-01
Abstract
We consider the Perona-Malik functional in dimension one, namely an integral functional whose Lagrangian is convex-concave with respect to the derivative, with a convexification that is identically zero. We approximate and regularize the functional by adding a term that depends on second order derivatives multiplied by a small coefficient.We investigate the asymptotic behavior of minima and minimizers as this small parameter vanishes. In particular, we show that minimizers exhibit the so-called staircasing phenomenon, namely they develop a sort of microstructure that looks like a piecewise constant function at a suitable scale.Our analysis relies on Gamma-convergence results for a rescaled functional, blow-up techniques, and a characterization of local minimizers for the limit problem. This approach can be extended to more general models.| File | Dimensione | Formato | |
|---|---|---|---|
|
TAMS2023.pdf
non disponibili
Tipologia:
Versione Editoriale (PDF)
Licenza:
Non specificato
Dimensione
687.72 kB
Formato
Adobe PDF
|
687.72 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
|
FastPM-CdV_v3.pdf
accesso aperto
Tipologia:
Documento in Pre-print
Licenza:
Non specificato
Dimensione
706.1 kB
Formato
Adobe PDF
|
706.1 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


