Scalable integrated region-based image retrieval using IRM and statistical clustering

James Wang, Yanping Du

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

3 Citations (Scopus)

Abstract

Statistical clustering is critical in designing scalable image retrieval systems. In this paper, we present a scalable algorithm for indexing and retrieving images based on region segmentation. The method uses statistical clustering on region features and IRM (Integrated Region Matching), a measure developed to evaluate overall similarity between images that incorporates properties of all the regions in the images by a region-matching scheme. Compared with retrieval based on individual regions, our overall similarity approach (a) reduces the influence of inaccurate segmentation, (b) helps to clarify the semantics of a particular region, and (c) enables a simple querying interface for region-based im- age retrieval systems. The algorithm has been implemented as a part of our experimental SIMPLI city image retrieval system and tested on large-scale image databases of both general-purpose images and pathology slides. Experiments have demonstrated that this technique maintains the accuracy and robustness of the original system while reducing the matching time significantly.

Original languageEnglish (US)
Title of host publicationProceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001
PublisherAssociation for Computing Machinery
Pages268-277
Number of pages10
ISBN (Print)1581133456, 9781581133455
DOIs
StatePublished - Jan 1 2001
Event1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001 - Roanoke, VA, United States
Duration: Jun 24 2001Jun 28 2001

Publication series

NameProceedings of the ACM International Conference on Digital Libraries

Other

Other1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001
CountryUnited States
CityRoanoke, VA
Period6/24/016/28/01

Fingerprint

Image retrieval
Pathology
Semantics
Experiments
indexing
pathology
semantics
experiment

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences

Cite this

Wang, J., & Du, Y. (2001). Scalable integrated region-based image retrieval using IRM and statistical clustering. In Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001 (pp. 268-277). (Proceedings of the ACM International Conference on Digital Libraries). Association for Computing Machinery. https://doi.org/10.1145/379437.379679
Wang, James ; Du, Yanping. / Scalable integrated region-based image retrieval using IRM and statistical clustering. Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001. Association for Computing Machinery, 2001. pp. 268-277 (Proceedings of the ACM International Conference on Digital Libraries).
@inproceedings{cb87b499c04c4467b6b6b2b60fe6e6c3,
title = "Scalable integrated region-based image retrieval using IRM and statistical clustering",
abstract = "Statistical clustering is critical in designing scalable image retrieval systems. In this paper, we present a scalable algorithm for indexing and retrieving images based on region segmentation. The method uses statistical clustering on region features and IRM (Integrated Region Matching), a measure developed to evaluate overall similarity between images that incorporates properties of all the regions in the images by a region-matching scheme. Compared with retrieval based on individual regions, our overall similarity approach (a) reduces the influence of inaccurate segmentation, (b) helps to clarify the semantics of a particular region, and (c) enables a simple querying interface for region-based im- age retrieval systems. The algorithm has been implemented as a part of our experimental SIMPLI city image retrieval system and tested on large-scale image databases of both general-purpose images and pathology slides. Experiments have demonstrated that this technique maintains the accuracy and robustness of the original system while reducing the matching time significantly.",
author = "James Wang and Yanping Du",
year = "2001",
month = "1",
day = "1",
doi = "10.1145/379437.379679",
language = "English (US)",
isbn = "1581133456",
series = "Proceedings of the ACM International Conference on Digital Libraries",
publisher = "Association for Computing Machinery",
pages = "268--277",
booktitle = "Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001",

}

Wang, J & Du, Y 2001, Scalable integrated region-based image retrieval using IRM and statistical clustering. in Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001. Proceedings of the ACM International Conference on Digital Libraries, Association for Computing Machinery, pp. 268-277, 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001, Roanoke, VA, United States, 6/24/01. https://doi.org/10.1145/379437.379679

Scalable integrated region-based image retrieval using IRM and statistical clustering. / Wang, James; Du, Yanping.

Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001. Association for Computing Machinery, 2001. p. 268-277 (Proceedings of the ACM International Conference on Digital Libraries).

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

TY - GEN

T1 - Scalable integrated region-based image retrieval using IRM and statistical clustering

AU - Wang, James

AU - Du, Yanping

PY - 2001/1/1

Y1 - 2001/1/1

N2 - Statistical clustering is critical in designing scalable image retrieval systems. In this paper, we present a scalable algorithm for indexing and retrieving images based on region segmentation. The method uses statistical clustering on region features and IRM (Integrated Region Matching), a measure developed to evaluate overall similarity between images that incorporates properties of all the regions in the images by a region-matching scheme. Compared with retrieval based on individual regions, our overall similarity approach (a) reduces the influence of inaccurate segmentation, (b) helps to clarify the semantics of a particular region, and (c) enables a simple querying interface for region-based im- age retrieval systems. The algorithm has been implemented as a part of our experimental SIMPLI city image retrieval system and tested on large-scale image databases of both general-purpose images and pathology slides. Experiments have demonstrated that this technique maintains the accuracy and robustness of the original system while reducing the matching time significantly.

AB - Statistical clustering is critical in designing scalable image retrieval systems. In this paper, we present a scalable algorithm for indexing and retrieving images based on region segmentation. The method uses statistical clustering on region features and IRM (Integrated Region Matching), a measure developed to evaluate overall similarity between images that incorporates properties of all the regions in the images by a region-matching scheme. Compared with retrieval based on individual regions, our overall similarity approach (a) reduces the influence of inaccurate segmentation, (b) helps to clarify the semantics of a particular region, and (c) enables a simple querying interface for region-based im- age retrieval systems. The algorithm has been implemented as a part of our experimental SIMPLI city image retrieval system and tested on large-scale image databases of both general-purpose images and pathology slides. Experiments have demonstrated that this technique maintains the accuracy and robustness of the original system while reducing the matching time significantly.

UR - http://www.scopus.com/inward/record.url?scp=84901285456&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901285456&partnerID=8YFLogxK

U2 - 10.1145/379437.379679

DO - 10.1145/379437.379679

M3 - Conference contribution

SN - 1581133456

SN - 9781581133455

T3 - Proceedings of the ACM International Conference on Digital Libraries

SP - 268

EP - 277

BT - Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001

PB - Association for Computing Machinery

ER -

Wang J, Du Y. Scalable integrated region-based image retrieval using IRM and statistical clustering. In Proceedings of the 1st ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL 2001. Association for Computing Machinery. 2001. p. 268-277. (Proceedings of the ACM International Conference on Digital Libraries). https://doi.org/10.1145/379437.379679