Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Similarity-based online feature selection in content-based image retrieval.

Wei Jiang1, Guihua Er, Qionghai Dai

  • 1Department of Automation, Tsinghua University, Beijing 100084, China. jiangwei98@mails.tsinghua.edu.cn

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|March 8, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Real-time robust autofocus method enabling sustained intravital scanning light field imaging.

Nature communications·2026
Same author

A multi-modal foundation model for brain disease diagnosis and medical imaging.

Patterns (New York, N.Y.)·2026
Same author

Modulation of place cells using targeted stimulation with bidirectional microelectrode arrays enhances spatial learning speed in mice.

Fundamental research·2026
Same author

Unsupervised transfer learning enables multi-animal tracking without training annotation.

Nature methods·2026
Same author

Dynamic neuronal ensembles encode burst-suppression revealed by cortex-wide optical-electrical interfaces.

Nature communications·2026
Same author

High-fidelity intravital imaging of biological dynamics with latent-space-enhanced digital adaptive optics.

Nature biotechnology·2026

This study introduces a novel online feature selection method for content-based image retrieval (CBIR) systems. The approach enhances region-based image retrieval performance by effectively bridging the gap between semantic concepts and visual features.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Information Retrieval

Background:

  • Content-based image retrieval (CBIR) performance is limited by the gap between high-level semantics and low-level visual features.
  • Online feature selection is crucial for bridging this semantic-visual gap in real-time systems.

Purpose of the Study:

  • To investigate online feature selection within relevance feedback for region-based image retrieval.
  • To improve the accuracy and efficiency of CBIR systems.

Main Methods:

  • A novel feature selection criterion based on psychological similarity between positive and negative training sets.
  • An online feature selection algorithm using a boosting approach to identify representative features.
  • A new region-based image representation in a uniform fuzzy feature space for CBIR.

Related Experiment Videos

Main Results:

  • The proposed method effectively improves retrieval performance in region-based image retrieval systems.
  • The algorithm demonstrates significant time savings during processing.
  • Experimental comparisons validate the effectiveness against state-of-the-art methods.

Conclusions:

  • The developed online feature selection method enhances CBIR systems by addressing the semantic-visual gap.
  • The approach is suitable for online relevance feedback, handling small training sets and asymmetric properties efficiently.
  • This work offers a promising direction for advancing region-based image retrieval technology.