In this work we investigate finite co-volume methods for solving Partial Differential Equation (PDE) based diffusion models for noise removal in functional surfaces. We generalized the model proposed by Tai et al. [1][2] based on the reconstruction of a noise-reduced surface from the smoothed normal field, considering a curvature preserving term. The discretization of the PDE model by basic finite co-volume schemes on unstructured grids is investigated. The accuracy of the numerical model is then improved by using an higher order optimal recovery based on Radial Basis Functions (RBF). Preliminary numerical results demonstrate the effectiveness of the new numerical approach. © Springer-Verlag Berlin Heidelberg 2007.
Morigi, S., Sgallari, F. (2007). An high order finite co-volume scheme for denoising using radial basis functions. Berlin : Springer Verlag [10.1007/978-3-540-72823-8_5].
An high order finite co-volume scheme for denoising using radial basis functions
Morigi S.;Sgallari F.
2007
Abstract
In this work we investigate finite co-volume methods for solving Partial Differential Equation (PDE) based diffusion models for noise removal in functional surfaces. We generalized the model proposed by Tai et al. [1][2] based on the reconstruction of a noise-reduced surface from the smoothed normal field, considering a curvature preserving term. The discretization of the PDE model by basic finite co-volume schemes on unstructured grids is investigated. The accuracy of the numerical model is then improved by using an higher order optimal recovery based on Radial Basis Functions (RBF). Preliminary numerical results demonstrate the effectiveness of the new numerical approach. © Springer-Verlag Berlin Heidelberg 2007.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.