Title :
Directional composite morphological filter in image processing
Author :
Li, Wei ; Ronsin, Joseph
Author_Institution :
INSA, Laboratoire ARTIST, 20 Avenue des Buttes de Coësmes 35043 RENNES Cedex, France
Abstract :
In this paper, two pairs of dual morphological filters, the composite morphological filters (CMFs) are introduced. They have distinctive properties comparing with other similar morphological operations. CMFs are used to construct a directional morphological filter in impulsive noise removal procedures. It is proven that it has better noise-removal and detail-preserving abilities than classical morphological filters and other non-linear filters such as median or centre weighted median based filters. A threshold scheme is added to improve the final filtering performances.
Keywords :
Digital filters; Filtering algorithms; Maximum likelihood detection; Noise; Noise measurement; Nonlinear filters;
Conference_Titel :
European Signal Processing Conference, 1996. EUSIPCO 1996. 8th
Conference_Location :
Trieste, Italy
Print_ISBN :
978-888-6179-83-6