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

Bayesian tree-structured image modeling using wavelet-domain hidden Markov models.

J K Romberg1, H Choi, R G Baraniuk

  • 1Department of Electrical and Computer Engineering, Rice University, Houston, TX 77005, USA. jrom@rice.edu

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|February 6, 2008
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

Influence of organic loading rate on bioflocculant production derived from waste glycerin pitch and mixed microbial culture in single-step system.

Bioresource technology·2026
Same author

<i>Notum</i> Regulates Tooth Root Morphogenesis by Modulating Wnt Signaling.

Journal of dental research·2025
Same author

Dermoscopy for lipidized dermatofibroma: A useful diagnostic tool.

Annales de dermatologie et de venereologie·2024
Same author

Low household income increases the risk of tuberculosis recurrence: a retrospective nationwide cohort study in South Korea.

Public health·2023
Same author

Tuberculosis and risk of Parkinson's disease: A nationwide cohort study.

Pulmonology·2022
Same author

Search for Dark Matter Axions with CAST-CAPP.

Nature communications·2022
Same journal

Style-Aware Contrastive Test-Time Adaptation: A Dual-Cache Model for Robust Vision-Language Alignment.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Semantic Frame Interpolation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Physics-Guided Cross-Modal Decoupling with Test-Time Adaptation for Hyperspectral Image Restoration.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

Simplified Hidden Markov Tree (HMT) models, including a universal HMT (uHMT), reduce computational costs for image processing. These models effectively capture image structure for tasks like denoising with minimal training.

Area of Science:

  • Signal Processing
  • Image Analysis
  • Statistical Modeling

Background:

  • Wavelet-domain Hidden Markov Models (HMT) are effective for statistical signal and image processing.
  • HMT models capture joint probability density of wavelet coefficients but require computationally intensive training.

Purpose of the Study:

  • To simplify the HMT framework by exploiting image self-similarity.
  • To introduce a Bayesian universal HMT (uHMT) that requires no training.
  • To evaluate the performance of simplified HMT models in image estimation and denoising tasks.

Main Methods:

  • Exploiting inherent self-similarity in real-world images to simplify HMT parameterization.
  • Developing a simplified HMT model with nine meta-parameters, independent of image size and wavelet scales.

Related Experiment Videos

  • Introducing a Bayesian universal HMT (uHMT) with fixed meta-parameters.
  • Conducting image estimation and denoising experiments.

    • Main Results:

    • Simplified HMT models retain nearly all key image structure compared to full HMT.
    • The universal HMT (uHMT) requires no training and demonstrates strong performance.
    • A fast shift-invariant HMT estimation algorithm outperforms existing wavelet-based estimators in visual quality and mean square error.

    Conclusions:

    • Simplified and universal HMT models offer a computationally efficient alternative to traditional HMTs for image processing.
    • These models maintain high fidelity in capturing image structures.
    • The proposed fast estimation algorithm enhances practical applicability in image denoising and estimation.