Properties of combinations of hierarchical watersheds

•Combinations of hierarchical watersheds are not hierarchical watersheds in general.•Hierarchical watersheds can be characterized through binary partition trees.•Some combinations result in flattened (simplified) hierarchical watersheds. In [19], the authors provide an evaluation of combinations of...

Full description

Saved in:
Bibliographic Details
Published in:Pattern recognition letters 2019-12, Vol.128, p.513-520
Main Authors: Santana Maia, Deise, Cousty, Jean, Najman, Laurent, Perret, Benjamin
Format: Article
Language:English
Subjects:
Computer Science
Evaluation
Flattening
Image Processing
Infimum
Watersheds
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Combinations of hierarchical watersheds are not hierarchical watersheds in general.•Hierarchical watersheds can be characterized through binary partition trees.•Some combinations result in flattened (simplified) hierarchical watersheds. In [19], the authors provide an evaluation of combinations of hierarchical watersheds, showing that some combinations outperform individual hierarchical watersheds. In this article, we aim to achieve a deeper understanding of those combinations. We study which of the four combinations evaluated in [19] always result in flattened (simplified) hierarchical watersheds. Then, we provide a sufficient condition for a combination to always output flattened hierarchical watersheds, and a new combining function that outputs flattened hierarchical watersheds. We conclude that, among the combinations assessed in [19], only a particular case of combinations with infimum always result in flattened hierarchical watersheds.
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2019.10.009