Vector image segmentation for content-based vector image retrieval

Takahiro Hayashi, Rikio Onai, Koji Abe

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

2 Citations (Scopus)

Abstract

This paper proposes a novel method for vector image segmentation as the first step in developing a content-based vector image retrieval system. Structure of a vector image can be represented as a tree in which each node is assigned each object region in the vector image and each link represents the inclusion relation between two object regions. In order to generate such trees, the method separates object regions from background by detecting figures defining boundary between object regions and background. The proposed method finds object regions before rasterizing, which is an essential difference from existing object separation methods. We have evaluated the effectiveness of the proposed object separation on 40 test vector images by comparing manual object separation. The experimental results have shown that the proposed method has a high performance which is comparable to manual object separation.

Original languageEnglish
Title of host publicationCIT 2007
Subtitle of host publication7th IEEE International Conference on Computer and Information Technology
Pages695-700
Number of pages6
DOIs
Publication statusPublished - 2007
Externally publishedYes
EventCIT 2007: 7th IEEE International Conference on Computer and Information Technology - Aizu-Wakamatsu, Fukushima
Duration: 2007 Oct 162007 Oct 19

Other

OtherCIT 2007: 7th IEEE International Conference on Computer and Information Technology
CityAizu-Wakamatsu, Fukushima
Period07/10/1607/10/19

Fingerprint

Image retrieval
Image Retrieval
Image segmentation
Image Segmentation
Object
Inclusion Relations
Figure
High Performance
Experimental Results
Vertex of a graph

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems
  • Software
  • Mathematics(all)

Cite this

Hayashi, T., Onai, R., & Abe, K. (2007). Vector image segmentation for content-based vector image retrieval. In CIT 2007: 7th IEEE International Conference on Computer and Information Technology (pp. 695-700). [4385166] https://doi.org/10.1109/CIT.2007.4385166

Vector image segmentation for content-based vector image retrieval. / Hayashi, Takahiro; Onai, Rikio; Abe, Koji.

CIT 2007: 7th IEEE International Conference on Computer and Information Technology. 2007. p. 695-700 4385166.

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

Hayashi, T, Onai, R & Abe, K 2007, Vector image segmentation for content-based vector image retrieval. in CIT 2007: 7th IEEE International Conference on Computer and Information Technology., 4385166, pp. 695-700, CIT 2007: 7th IEEE International Conference on Computer and Information Technology, Aizu-Wakamatsu, Fukushima, 07/10/16. https://doi.org/10.1109/CIT.2007.4385166
Hayashi T, Onai R, Abe K. Vector image segmentation for content-based vector image retrieval. In CIT 2007: 7th IEEE International Conference on Computer and Information Technology. 2007. p. 695-700. 4385166 https://doi.org/10.1109/CIT.2007.4385166
Hayashi, Takahiro ; Onai, Rikio ; Abe, Koji. / Vector image segmentation for content-based vector image retrieval. CIT 2007: 7th IEEE International Conference on Computer and Information Technology. 2007. pp. 695-700
@inproceedings{9eedd22841404f57949578410dbd020c,
title = "Vector image segmentation for content-based vector image retrieval",
abstract = "This paper proposes a novel method for vector image segmentation as the first step in developing a content-based vector image retrieval system. Structure of a vector image can be represented as a tree in which each node is assigned each object region in the vector image and each link represents the inclusion relation between two object regions. In order to generate such trees, the method separates object regions from background by detecting figures defining boundary between object regions and background. The proposed method finds object regions before rasterizing, which is an essential difference from existing object separation methods. We have evaluated the effectiveness of the proposed object separation on 40 test vector images by comparing manual object separation. The experimental results have shown that the proposed method has a high performance which is comparable to manual object separation.",
author = "Takahiro Hayashi and Rikio Onai and Koji Abe",
year = "2007",
doi = "10.1109/CIT.2007.4385166",
language = "English",
isbn = "0769529836",
pages = "695--700",
booktitle = "CIT 2007",

}

TY - GEN

T1 - Vector image segmentation for content-based vector image retrieval

AU - Hayashi, Takahiro

AU - Onai, Rikio

AU - Abe, Koji

PY - 2007

Y1 - 2007

N2 - This paper proposes a novel method for vector image segmentation as the first step in developing a content-based vector image retrieval system. Structure of a vector image can be represented as a tree in which each node is assigned each object region in the vector image and each link represents the inclusion relation between two object regions. In order to generate such trees, the method separates object regions from background by detecting figures defining boundary between object regions and background. The proposed method finds object regions before rasterizing, which is an essential difference from existing object separation methods. We have evaluated the effectiveness of the proposed object separation on 40 test vector images by comparing manual object separation. The experimental results have shown that the proposed method has a high performance which is comparable to manual object separation.

AB - This paper proposes a novel method for vector image segmentation as the first step in developing a content-based vector image retrieval system. Structure of a vector image can be represented as a tree in which each node is assigned each object region in the vector image and each link represents the inclusion relation between two object regions. In order to generate such trees, the method separates object regions from background by detecting figures defining boundary between object regions and background. The proposed method finds object regions before rasterizing, which is an essential difference from existing object separation methods. We have evaluated the effectiveness of the proposed object separation on 40 test vector images by comparing manual object separation. The experimental results have shown that the proposed method has a high performance which is comparable to manual object separation.

UR - http://www.scopus.com/inward/record.url?scp=38049023989&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38049023989&partnerID=8YFLogxK

U2 - 10.1109/CIT.2007.4385166

DO - 10.1109/CIT.2007.4385166

M3 - Conference contribution

SN - 0769529836

SN - 9780769529837

SP - 695

EP - 700

BT - CIT 2007

ER -