Probabilistic semantic network-based image retrieval using MMM and relevance feedback

Mei-Ling Shyu, Shu Ching Chen, Min Chen, Chengcui Zhang, Chi Min Shu

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The performance of content-based image retrieval (CBIR) systems is largely limited by the gap between the low-level features and high-level semantic concepts. In this paper, a probabilistic semantic network-based image retrieval framework using relevance feedback is proposed to bridge this gap, which not only takes into consideration the low-level image content features, but also learns high-level concepts from a set of training data, such as access frequencies and access patterns of the images. One of the distinct properties of our framework is that it exploits the structured description of visual contents as well as the relative affinity measurements among the images. Consequently, it provides the capability to bridge the gap between the low-level features and high-level concepts. Moreover, such high-level concepts can be learned off-line, and can be utilized and refined based on the user's specific interest during the on-line retrieval process. Our experimental results demonstrate that the proposed framework can effectively assist in retrieving more accurate results for user queries.

Original languageEnglish
Pages (from-to)131-147
Number of pages17
JournalMultimedia Tools and Applications
Volume30
Issue number2
DOIs
StatePublished - Aug 1 2006

Fingerprint

Semantic Network
Relevance Feedback
Image retrieval
Image Retrieval
Semantics
Feedback
Content-based Image Retrieval
Affine transformation
Retrieval
Query
Distinct
Concepts
Line
Experimental Results
Demonstrate
Framework

Keywords

  • Content-based image retrieval
  • MMM mechanism
  • Probabilistic semantic network
  • Relevance feedback

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

Probabilistic semantic network-based image retrieval using MMM and relevance feedback. / Shyu, Mei-Ling; Chen, Shu Ching; Chen, Min; Zhang, Chengcui; Shu, Chi Min.

In: Multimedia Tools and Applications, Vol. 30, No. 2, 01.08.2006, p. 131-147.

Research output: Contribution to journalArticle

Shyu, Mei-Ling ; Chen, Shu Ching ; Chen, Min ; Zhang, Chengcui ; Shu, Chi Min. / Probabilistic semantic network-based image retrieval using MMM and relevance feedback. In: Multimedia Tools and Applications. 2006 ; Vol. 30, No. 2. pp. 131-147.
@article{1e30841516a243fe88093366ca4940f5,
title = "Probabilistic semantic network-based image retrieval using MMM and relevance feedback",
abstract = "The performance of content-based image retrieval (CBIR) systems is largely limited by the gap between the low-level features and high-level semantic concepts. In this paper, a probabilistic semantic network-based image retrieval framework using relevance feedback is proposed to bridge this gap, which not only takes into consideration the low-level image content features, but also learns high-level concepts from a set of training data, such as access frequencies and access patterns of the images. One of the distinct properties of our framework is that it exploits the structured description of visual contents as well as the relative affinity measurements among the images. Consequently, it provides the capability to bridge the gap between the low-level features and high-level concepts. Moreover, such high-level concepts can be learned off-line, and can be utilized and refined based on the user's specific interest during the on-line retrieval process. Our experimental results demonstrate that the proposed framework can effectively assist in retrieving more accurate results for user queries.",
keywords = "Content-based image retrieval, MMM mechanism, Probabilistic semantic network, Relevance feedback",
author = "Mei-Ling Shyu and Chen, {Shu Ching} and Min Chen and Chengcui Zhang and Shu, {Chi Min}",
year = "2006",
month = "8",
day = "1",
doi = "10.1007/s11042-006-0023-5",
language = "English",
volume = "30",
pages = "131--147",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",
number = "2",

}

TY - JOUR

T1 - Probabilistic semantic network-based image retrieval using MMM and relevance feedback

AU - Shyu, Mei-Ling

AU - Chen, Shu Ching

AU - Chen, Min

AU - Zhang, Chengcui

AU - Shu, Chi Min

PY - 2006/8/1

Y1 - 2006/8/1

N2 - The performance of content-based image retrieval (CBIR) systems is largely limited by the gap between the low-level features and high-level semantic concepts. In this paper, a probabilistic semantic network-based image retrieval framework using relevance feedback is proposed to bridge this gap, which not only takes into consideration the low-level image content features, but also learns high-level concepts from a set of training data, such as access frequencies and access patterns of the images. One of the distinct properties of our framework is that it exploits the structured description of visual contents as well as the relative affinity measurements among the images. Consequently, it provides the capability to bridge the gap between the low-level features and high-level concepts. Moreover, such high-level concepts can be learned off-line, and can be utilized and refined based on the user's specific interest during the on-line retrieval process. Our experimental results demonstrate that the proposed framework can effectively assist in retrieving more accurate results for user queries.

AB - The performance of content-based image retrieval (CBIR) systems is largely limited by the gap between the low-level features and high-level semantic concepts. In this paper, a probabilistic semantic network-based image retrieval framework using relevance feedback is proposed to bridge this gap, which not only takes into consideration the low-level image content features, but also learns high-level concepts from a set of training data, such as access frequencies and access patterns of the images. One of the distinct properties of our framework is that it exploits the structured description of visual contents as well as the relative affinity measurements among the images. Consequently, it provides the capability to bridge the gap between the low-level features and high-level concepts. Moreover, such high-level concepts can be learned off-line, and can be utilized and refined based on the user's specific interest during the on-line retrieval process. Our experimental results demonstrate that the proposed framework can effectively assist in retrieving more accurate results for user queries.

KW - Content-based image retrieval

KW - MMM mechanism

KW - Probabilistic semantic network

KW - Relevance feedback

UR - http://www.scopus.com/inward/record.url?scp=33748494007&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748494007&partnerID=8YFLogxK

U2 - 10.1007/s11042-006-0023-5

DO - 10.1007/s11042-006-0023-5

M3 - Article

AN - SCOPUS:33748494007

VL - 30

SP - 131

EP - 147

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 2

ER -