Building a navigation structure from a fuzzy relationship for image retrieval

Erwan Loisant, Hiroshi Ishikawa, José Martinez, Manabu Ohta, Kaoru Katayama

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

Abstract

Going one step further feedback querying in integrating user into retrieval process, navigation is the more recent approach to find images in a large image collection using content-based information. However, while properties extracted from images are usually fuzzy data, most of the time a navigation structure will deal with binary links from an image (or a group of images) to another. A trivial solution to get a binary relationship from fuzzy data is to apply a threshold, but this solution not only leads to a loss of information but fails to distinguish noise from interesting elements. In this paper, we propose two techniques to eliminate isolated elements and lead to a structure made of more compact subparts. The first one is based on a variable threshold depending on the number of neighbours. The second one, specific to Galois' lattice, is based on taking into account the existing navigation structure for binarisation of descriptions. Experiments showed that it improves the resulting structure by reducing the number of nodes without loosing information on image description, thus improving user experience.

Original languageEnglish
Title of host publicationProceedings - International Workshop on Biomedical Data Engineering, BMDE2005
Volume2005
DOIs
Publication statusPublished - 2005
Externally publishedYes
Event21st International Conference on Data Engineering Workshops 2005 - Tokyo, Japan
Duration: Apr 3 2005Apr 4 2005

Other

Other21st International Conference on Data Engineering Workshops 2005
CountryJapan
CityTokyo
Period4/3/054/4/05

Fingerprint

Image retrieval
Navigation
Feedback
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Loisant, E., Ishikawa, H., Martinez, J., Ohta, M., & Katayama, K. (2005). Building a navigation structure from a fuzzy relationship for image retrieval. In Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005 (Vol. 2005). [1647785] https://doi.org/10.1109/ICDE.2005.201

Building a navigation structure from a fuzzy relationship for image retrieval. / Loisant, Erwan; Ishikawa, Hiroshi; Martinez, José; Ohta, Manabu; Katayama, Kaoru.

Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005. Vol. 2005 2005. 1647785.

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

Loisant, E, Ishikawa, H, Martinez, J, Ohta, M & Katayama, K 2005, Building a navigation structure from a fuzzy relationship for image retrieval. in Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005. vol. 2005, 1647785, 21st International Conference on Data Engineering Workshops 2005, Tokyo, Japan, 4/3/05. https://doi.org/10.1109/ICDE.2005.201
Loisant E, Ishikawa H, Martinez J, Ohta M, Katayama K. Building a navigation structure from a fuzzy relationship for image retrieval. In Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005. Vol. 2005. 2005. 1647785 https://doi.org/10.1109/ICDE.2005.201
Loisant, Erwan ; Ishikawa, Hiroshi ; Martinez, José ; Ohta, Manabu ; Katayama, Kaoru. / Building a navigation structure from a fuzzy relationship for image retrieval. Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005. Vol. 2005 2005.
@inproceedings{26b594f955e645a4975799b5e3f00319,
title = "Building a navigation structure from a fuzzy relationship for image retrieval",
abstract = "Going one step further feedback querying in integrating user into retrieval process, navigation is the more recent approach to find images in a large image collection using content-based information. However, while properties extracted from images are usually fuzzy data, most of the time a navigation structure will deal with binary links from an image (or a group of images) to another. A trivial solution to get a binary relationship from fuzzy data is to apply a threshold, but this solution not only leads to a loss of information but fails to distinguish noise from interesting elements. In this paper, we propose two techniques to eliminate isolated elements and lead to a structure made of more compact subparts. The first one is based on a variable threshold depending on the number of neighbours. The second one, specific to Galois' lattice, is based on taking into account the existing navigation structure for binarisation of descriptions. Experiments showed that it improves the resulting structure by reducing the number of nodes without loosing information on image description, thus improving user experience.",
author = "Erwan Loisant and Hiroshi Ishikawa and Jos{\'e} Martinez and Manabu Ohta and Kaoru Katayama",
year = "2005",
doi = "10.1109/ICDE.2005.201",
language = "English",
isbn = "0769526578",
volume = "2005",
booktitle = "Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005",

}

TY - GEN

T1 - Building a navigation structure from a fuzzy relationship for image retrieval

AU - Loisant, Erwan

AU - Ishikawa, Hiroshi

AU - Martinez, José

AU - Ohta, Manabu

AU - Katayama, Kaoru

PY - 2005

Y1 - 2005

N2 - Going one step further feedback querying in integrating user into retrieval process, navigation is the more recent approach to find images in a large image collection using content-based information. However, while properties extracted from images are usually fuzzy data, most of the time a navigation structure will deal with binary links from an image (or a group of images) to another. A trivial solution to get a binary relationship from fuzzy data is to apply a threshold, but this solution not only leads to a loss of information but fails to distinguish noise from interesting elements. In this paper, we propose two techniques to eliminate isolated elements and lead to a structure made of more compact subparts. The first one is based on a variable threshold depending on the number of neighbours. The second one, specific to Galois' lattice, is based on taking into account the existing navigation structure for binarisation of descriptions. Experiments showed that it improves the resulting structure by reducing the number of nodes without loosing information on image description, thus improving user experience.

AB - Going one step further feedback querying in integrating user into retrieval process, navigation is the more recent approach to find images in a large image collection using content-based information. However, while properties extracted from images are usually fuzzy data, most of the time a navigation structure will deal with binary links from an image (or a group of images) to another. A trivial solution to get a binary relationship from fuzzy data is to apply a threshold, but this solution not only leads to a loss of information but fails to distinguish noise from interesting elements. In this paper, we propose two techniques to eliminate isolated elements and lead to a structure made of more compact subparts. The first one is based on a variable threshold depending on the number of neighbours. The second one, specific to Galois' lattice, is based on taking into account the existing navigation structure for binarisation of descriptions. Experiments showed that it improves the resulting structure by reducing the number of nodes without loosing information on image description, thus improving user experience.

UR - http://www.scopus.com/inward/record.url?scp=33947158150&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947158150&partnerID=8YFLogxK

U2 - 10.1109/ICDE.2005.201

DO - 10.1109/ICDE.2005.201

M3 - Conference contribution

SN - 0769526578

SN - 9780769526577

VL - 2005

BT - Proceedings - International Workshop on Biomedical Data Engineering, BMDE2005

ER -