Lie group of transformations to study anisotropic diffusion of scalar images (Q1933377)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Lie group of transformations to study anisotropic diffusion of scalar images |
scientific article |
Statements
Lie group of transformations to study anisotropic diffusion of scalar images (English)
0 references
23 January 2013
0 references
A Lie group of transformations associated with an anisotropic diffusion equation is derived using invariance properties. The equation is of great importance in the study of the geometrical diffusion of scalar images. The author employs the method of Lie group of transformations after transforming the equation to a system of first-order partial differential equations. Constant conformal invariance of the considered system of equations is used to obtain a system of equations for the evaluation of the infinitesimals. Particular cases are considered by asigning particular values to the arbitrary function constants associated with the infinitesimals. It was observed that the edge stopping functions used by Peron and Malik can be obtained in particular cases. In view of the assumptions, the number of independent variables in the evolution equation are reduced, which is useful for the numerical implementation in the study of the anisotropic diffusion of scaler images. Infinitesimals of the Lie group of transformations with respect to the independent and dependent variables along with invariant surface conditions are used to obtain a general form of edge stopping function. The edge stopping functions used by Perona and Malik and their successors are obtained as particular cases. The equation in three independent variables is reduced in terms of fewer independent variables.
0 references
Lie group of transformations
0 references
partial differential equations
0 references
self-similar solutions
0 references
image processing
0 references
edge stopping function
0 references
0 references
0 references