scientific article; zbMATH DE number 1125941
The multi-scale analysis can be the basis for efficient algorithms for denoising images and edge-detection. Diffusion processes create multi-scale analyses which enable the generation of simplified pictures, where for increasing scale the image gets sketchier. In many practical applications the ``scaled image can be characterized by a variational formulation as the solution of a minimization problem involving unbounded operators. The concept of stable evaluation of unbounded operators is outlined and it is shown that this concept is relevant for multi-scale analysis associated with diffusion processes. The unbounded operators can be evaluated by regularization techniques. Numerical experiments are presented which show that regularization with higher-order bounded variation resolves discontinuities in derivatives efficiently. In comparison with a linear multi-scale algorithm the location of the discontinuities can be sharply estimated. On the other side the amount of computations is increasing in the case of the nonlinear multi-scale filtering.
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4379832)