Unsupervised image segmentation using comparative reasoning and random walks

Anuva Kulkarni, Filipe Condessa, Jelena Kovacevic

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

Abstract

An image segmentation method that does not need training data can provide faster results than methods using complex optimization. Motivated by this idea, we present an unsupervised image segmentation method that combines comparative reasoning with graph-based clustering. Comparative reasoning enables fast similarity search on the image, and these search results are used with the Random Walks algorithm, which is used for clustering and calculating class probabilities. Our method is validated on diverse image modalities such as biomedical images, natural images and texture images. The performance of the method is measured through cluster purity based on available ground truth. Our results are compared to existing segmentation methods using Global Consistency Error scores.

Original languageEnglish (US)
Title of host publication2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages338-342
Number of pages5
ISBN (Electronic)9781479975914
DOIs
StatePublished - Feb 23 2016
EventIEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 - Orlando, United States
Duration: Dec 13 2015Dec 16 2015

Other

OtherIEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
CountryUnited States
CityOrlando
Period12/13/1512/16/15

Fingerprint

Image segmentation
Textures

Keywords

  • comparative reasoning
  • hashing
  • Random Walks
  • Unsupervised image segmentation
  • Winner Take All hash

ASJC Scopus subject areas

  • Information Systems
  • Signal Processing

Cite this

Kulkarni, A., Condessa, F., & Kovacevic, J. (2016). Unsupervised image segmentation using comparative reasoning and random walks. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 (pp. 338-342). [7418213] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2015.7418213

Unsupervised image segmentation using comparative reasoning and random walks. / Kulkarni, Anuva; Condessa, Filipe; Kovacevic, Jelena.

2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 338-342 7418213.

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

Kulkarni, A, Condessa, F & Kovacevic, J 2016, Unsupervised image segmentation using comparative reasoning and random walks. in 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015., 7418213, Institute of Electrical and Electronics Engineers Inc., pp. 338-342, IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015, Orlando, United States, 12/13/15. https://doi.org/10.1109/GlobalSIP.2015.7418213
Kulkarni A, Condessa F, Kovacevic J. Unsupervised image segmentation using comparative reasoning and random walks. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 338-342. 7418213 https://doi.org/10.1109/GlobalSIP.2015.7418213
Kulkarni, Anuva ; Condessa, Filipe ; Kovacevic, Jelena. / Unsupervised image segmentation using comparative reasoning and random walks. 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 338-342
@inproceedings{534d7ae6c0ea45d2ad52e6cf612aeb7f,
title = "Unsupervised image segmentation using comparative reasoning and random walks",
abstract = "An image segmentation method that does not need training data can provide faster results than methods using complex optimization. Motivated by this idea, we present an unsupervised image segmentation method that combines comparative reasoning with graph-based clustering. Comparative reasoning enables fast similarity search on the image, and these search results are used with the Random Walks algorithm, which is used for clustering and calculating class probabilities. Our method is validated on diverse image modalities such as biomedical images, natural images and texture images. The performance of the method is measured through cluster purity based on available ground truth. Our results are compared to existing segmentation methods using Global Consistency Error scores.",
keywords = "comparative reasoning, hashing, Random Walks, Unsupervised image segmentation, Winner Take All hash",
author = "Anuva Kulkarni and Filipe Condessa and Jelena Kovacevic",
year = "2016",
month = "2",
day = "23",
doi = "10.1109/GlobalSIP.2015.7418213",
language = "English (US)",
pages = "338--342",
booktitle = "2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Unsupervised image segmentation using comparative reasoning and random walks

AU - Kulkarni, Anuva

AU - Condessa, Filipe

AU - Kovacevic, Jelena

PY - 2016/2/23

Y1 - 2016/2/23

N2 - An image segmentation method that does not need training data can provide faster results than methods using complex optimization. Motivated by this idea, we present an unsupervised image segmentation method that combines comparative reasoning with graph-based clustering. Comparative reasoning enables fast similarity search on the image, and these search results are used with the Random Walks algorithm, which is used for clustering and calculating class probabilities. Our method is validated on diverse image modalities such as biomedical images, natural images and texture images. The performance of the method is measured through cluster purity based on available ground truth. Our results are compared to existing segmentation methods using Global Consistency Error scores.

AB - An image segmentation method that does not need training data can provide faster results than methods using complex optimization. Motivated by this idea, we present an unsupervised image segmentation method that combines comparative reasoning with graph-based clustering. Comparative reasoning enables fast similarity search on the image, and these search results are used with the Random Walks algorithm, which is used for clustering and calculating class probabilities. Our method is validated on diverse image modalities such as biomedical images, natural images and texture images. The performance of the method is measured through cluster purity based on available ground truth. Our results are compared to existing segmentation methods using Global Consistency Error scores.

KW - comparative reasoning

KW - hashing

KW - Random Walks

KW - Unsupervised image segmentation

KW - Winner Take All hash

UR - http://www.scopus.com/inward/record.url?scp=84964746660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964746660&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2015.7418213

DO - 10.1109/GlobalSIP.2015.7418213

M3 - Conference contribution

SP - 338

EP - 342

BT - 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -