Title :
Adaptive directional morphological operators
Author :
Terebes, Romulus ; Lavialle, Olivier ; Baylou, Pierre ; Borda, Monica ; Nafornita, Ioan
Author_Institution :
Equipe Signal/Image ENSERB et GDR-ISIS-CNRS, Av. du Dr. Schweitzer, BP 99, 33402 Talence Cedex, France
Abstract :
This paper proposes a new method of nonlinear filtering based on the classical morphological operators. The originality of our method consists in assigning to each pixel an adaptive structuring element depending on the local orientation of the image. Several examples of applications are presented and a comparison with the classical operators is done. Our method can be applied for early vision tasks.
Keywords :
Filtering; Image reconstruction; Morphological operations; Morphology; Principal component analysis; Transforms; Vectors;
Conference_Titel :
Signal Processing Conference, 2000 10th European
Conference_Location :
Tampere, Finland
Print_ISBN :
978-952-1504-43-3
