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

Rotation-invariant texture retrieval with Gaussianized steerable pyramids.

George Tzagkarakis1, Baltasar Beferull-Lozano, Panagiotis Tsakalides

  • 1Department of Computer Science, University of Crete and Institute of Computer Science (ICS-FORTH), 711 10 Heraklion, Crete, Greece. gtzag@ics.forth.gr

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|September 5, 2006
PubMed
Summary

This study introduces a new method for image retrieval that is invariant to rotation. It uses a steerable pyramid to analyze texture, making it easier to find similar images regardless of their orientation.

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

Classifier-driven generative adversarial networks for enhanced antimicrobial peptide design.

Briefings in bioinformatics·2025
Same author

Recurrence Quantification Analysis for Scene Change Detection and Foreground/Background Segmentation in Videos.

Journal of imaging·2025
Same author

De Novo Antimicrobial Peptide Design with Feedback Generative Adversarial Networks.

International journal of molecular sciences·2024
Same author

A Generative Neighborhood-Based Deep Autoencoder for Robust Imbalanced Classification.

IEEE transactions on artificial intelligence·2024
Same author

Embedded Temporal Convolutional Networks for Essential Climate Variables Forecasting.

Sensors (Basel, Switzerland)·2022
Same author

Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks.

Journal of imaging·2021

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Texture analysis is crucial for image retrieval.
  • Existing methods often struggle with rotation invariance.
  • Non-Gaussian distributions are common in image subband coefficients.

Purpose of the Study:

  • To develop a rotation-invariant image retrieval scheme.
  • To effectively capture non-Gaussian texture information.
  • To improve the accuracy of texture-based image similarity measurement.

Main Methods:

  • Utilizing a steerable pyramid for texture transformation.
  • Fitting subband coefficients with a joint alpha-stable sub-Gaussian model.
  • Applying normalization to Gaussianize coefficients.

Related Experiment Videos

  • Estimating covariances of normalized coefficients for feature extraction.
  • Employing a rotation-invariant Kullback-Leibler Divergence for similarity measurement.

    • Main Results:

    • The proposed method achieves rotation invariance in texture-based image retrieval.
    • The joint alpha-stable sub-Gaussian model effectively captures non-Gaussian texture properties.
    • Normalization successfully Gaussianizes coefficients for robust feature extraction.
    • Minimizing Kullback-Leibler Divergence over rotation angles provides accurate similarity measures.

    Conclusions:

    • The novel scheme offers robust rotation-invariant image retrieval.
    • The combination of steerable pyramid and statistical modeling enhances texture analysis.
    • This approach advances the field of content-based image retrieval, particularly for textured images.