Random walker watersheds

A new image segmentation approach

Sundaresh Ram, Jeffrey J Rodriguez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

We propose a new graph-based approach for performing a multilabel, interactive image segmentation using the principle of random walks. Using the random walk principle, given a set of user-defined (or prelabeled) pixels as labels, one can analytically calculate the probability of walking from each unlabeled pixel to each labeled pixel, thereby defining a vector of probabilities for each unlabeled pixel. By efficiently combining this vector of probabilities obtained for each unlabeled pixel, they can be assigned to one of the labels using the watershed algorithm to obtain an image segmentation. We present quantitative and qualitative results, comparing our new algorithm with the original random walker image segmentation algorithm.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages1473-1477
Number of pages5
DOIs
StatePublished - Oct 18 2013
Event2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Vancouver, BC, Canada
Duration: May 26 2013May 31 2013

Other

Other2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
CountryCanada
CityVancouver, BC
Period5/26/135/31/13

Fingerprint

Watersheds
Image segmentation
Pixels
Labels

Keywords

  • combinatorial Dirichlet problem
  • graph theory
  • Image segmentation
  • random walks
  • watersheds

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Ram, S., & Rodriguez, J. J. (2013). Random walker watersheds: A new image segmentation approach. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 1473-1477). [6637896] https://doi.org/10.1109/ICASSP.2013.6637896

Random walker watersheds : A new image segmentation approach. / Ram, Sundaresh; Rodriguez, Jeffrey J.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. p. 1473-1477 6637896.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ram, S & Rodriguez, JJ 2013, Random walker watersheds: A new image segmentation approach. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 6637896, pp. 1473-1477, 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013, Vancouver, BC, Canada, 5/26/13. https://doi.org/10.1109/ICASSP.2013.6637896
Ram S, Rodriguez JJ. Random walker watersheds: A new image segmentation approach. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. p. 1473-1477. 6637896 https://doi.org/10.1109/ICASSP.2013.6637896
Ram, Sundaresh ; Rodriguez, Jeffrey J. / Random walker watersheds : A new image segmentation approach. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. pp. 1473-1477
@inproceedings{233cd5e6b6864bd28758bba784eb2462,
title = "Random walker watersheds: A new image segmentation approach",
abstract = "We propose a new graph-based approach for performing a multilabel, interactive image segmentation using the principle of random walks. Using the random walk principle, given a set of user-defined (or prelabeled) pixels as labels, one can analytically calculate the probability of walking from each unlabeled pixel to each labeled pixel, thereby defining a vector of probabilities for each unlabeled pixel. By efficiently combining this vector of probabilities obtained for each unlabeled pixel, they can be assigned to one of the labels using the watershed algorithm to obtain an image segmentation. We present quantitative and qualitative results, comparing our new algorithm with the original random walker image segmentation algorithm.",
keywords = "combinatorial Dirichlet problem, graph theory, Image segmentation, random walks, watersheds",
author = "Sundaresh Ram and Rodriguez, {Jeffrey J}",
year = "2013",
month = "10",
day = "18",
doi = "10.1109/ICASSP.2013.6637896",
language = "English (US)",
isbn = "9781479903566",
pages = "1473--1477",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Random walker watersheds

T2 - A new image segmentation approach

AU - Ram, Sundaresh

AU - Rodriguez, Jeffrey J

PY - 2013/10/18

Y1 - 2013/10/18

N2 - We propose a new graph-based approach for performing a multilabel, interactive image segmentation using the principle of random walks. Using the random walk principle, given a set of user-defined (or prelabeled) pixels as labels, one can analytically calculate the probability of walking from each unlabeled pixel to each labeled pixel, thereby defining a vector of probabilities for each unlabeled pixel. By efficiently combining this vector of probabilities obtained for each unlabeled pixel, they can be assigned to one of the labels using the watershed algorithm to obtain an image segmentation. We present quantitative and qualitative results, comparing our new algorithm with the original random walker image segmentation algorithm.

AB - We propose a new graph-based approach for performing a multilabel, interactive image segmentation using the principle of random walks. Using the random walk principle, given a set of user-defined (or prelabeled) pixels as labels, one can analytically calculate the probability of walking from each unlabeled pixel to each labeled pixel, thereby defining a vector of probabilities for each unlabeled pixel. By efficiently combining this vector of probabilities obtained for each unlabeled pixel, they can be assigned to one of the labels using the watershed algorithm to obtain an image segmentation. We present quantitative and qualitative results, comparing our new algorithm with the original random walker image segmentation algorithm.

KW - combinatorial Dirichlet problem

KW - graph theory

KW - Image segmentation

KW - random walks

KW - watersheds

UR - http://www.scopus.com/inward/record.url?scp=84890499772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890499772&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2013.6637896

DO - 10.1109/ICASSP.2013.6637896

M3 - Conference contribution

SN - 9781479903566

SP - 1473

EP - 1477

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -