DMCA
Power Watershed: A Unifying Graph-Based Optimization Framework (2011)
Cached
Download Links
| Citations: | 42 - 8 self |
BibTeX
@MISC{Couprie11powerwatershed:,
author = {Camille Couprie and Leo Grady and Laurent Najman and Hugues Talbot},
title = {Power Watershed: A Unifying Graph-Based Optimization Framework},
year = {2011}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
In this work, we extend a common framework for graph-based image segmentation that includes the graph cuts, random walker, and shortest path optimization algorithms. Viewing an image as a weighted graph, these algorithms can be expressed by means of a common energy function with differing choices of a parameter q acting as an exponent on the differences between neighboring nodes. Introducing a new parameter p that fixes a power for the edge weights allows us to also include the optimal spanning forest algorithm for watershed in this same framework. We then propose a new family of segmentation algorithms that fixes p to produce an optimal spanning forest but varies the power q beyond the usual watershed algorithm, which we term power watershed. In particular when q = 2, the power watershed leads to a multilabel, scale and contrast invariant, unique global optimum obtained in practice in quasi-linear time. Placing the watershed algorithm in this energy minimization framework also opens new possibilities for using unary terms in traditional watershed segmentation and using watershed to optimize more general models of use in applications beyond image segmentation.
Keyphrases
graph-based optimization framework optimal spanning new parameter image segmentation new possibility edge weight unique global optimum term power path optimization algorithm energy minimization framework new family graph-based image segmentation graph cut unary term contrast invariant segmentation algorithm watershed algorithm random walker general model usual watershed algorithm weighted graph traditional watershed segmentation common energy function quasi-linear time common framework
