Morphology on color images
From LRDE
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.
- Authors
- Alexandre Abraham
- Type
- techreport
- Year
- 2009
- Number
- 0835
- Projects
- Olena
- Keywords
- watershed image segmentation, Color, MorphologyMultivaluate Ordering, Vector Median Filter, Autarkical Leveling, Color, Morphology, Multivaluate Ordering, Vector Median Filter, Autarkical Leveling
Abstract
This work takes place in the context of Milena, the C++ generic image processing library of the Olena platformdeveloped at LRDE. Morphological algorithms are one of Milena's major assets. Yet few image processing libraries implement them even though they are very useful. Those algorithms require supremum and infimum operators, which don't exist by default for composite types like red-green-blue (RGB) pixels. We therefore propose the implementation of those operators for RGB values, along with a complete toolchain allowing morphological algorithms to work on color images.