Leandro Coser,  Antonio Carlos Sobieranski,  Eros Comunello,  Aldo von Wangenheim. 

Anisotropic diffusion filtering is a well-established technique for image enhancement that works by means of diffusion functions. They are able to smooth images without destroying edge information. However, when many filtering iterations are applied or higher contrast parameter are used, edges gradually fade away and are ultimately smoothed by the process. We propose the adoption of a color gradient map and an adaptive contrast parameter to guide the smoothing in order to preserve the edges even after many iterations. Preliminary experiments show good results when compared with the traditional anisotropic diffusion filter.

http://www.lbd.dcc.ufmg.br/colecoes/colibri/2009/035.pdf

Processando consulta.