TY  - CONF
AU  - Kleefeld, Andreas
TI  - Mathematical morphology for multispectral images
M1  - FZJ-2018-03426
PY  - 2018
AB  - The processing of multispectral images is a challenging task due to the fact that the pixel content is vectorial and the definition of an order is difficult. A new geometrical framework based on double hypersimplices and the Loewner ordering is illustrated to define the two fundamental operations dilation and erosion for multispectral images. These are the two main building blocks of mathematical morphology to define higher morphological operations such as top hats, gradients, and the morphological Laplacian. Numerical results are given to show the advantages and shortcomings of the new proposed approach.
T2  - SIAM Conference on Imaging Science 2018
CY  - 5 Jun 2018 - 8 Jun 2018, Bologna (Italy)
Y2  - 5 Jun 2018 - 8 Jun 2018
M2  - Bologna, Italy
LB  - PUB:(DE-HGF)6
UR  - https://juser.fz-juelich.de/record/848160
ER  -