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

Regression-based prediction for blocking artifact reduction in JPEG-compressed images.

Kiryung Lee1, Dong Sik Kim, Taejeong Kim

  • 1Electronics and Telecommunications Research Institute, Daejeon, Korea. kiryung@etri.re.kr

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|January 14, 2005
PubMed
Summary

This study introduces a novel noniterative algorithm to reduce blocking artifacts in Joint Photographic Experts Group (JPEG) compressed images. The method uses low-pass filtering and adaptive prediction for improved image quality with less computational cost.

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

Feasibility Study on Bone Mineral Density Measurement Based on Dual-Layer Flat Panel Detector.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Performance of Dual-Layer Flat-Panel Detectors.

Diagnostics (Basel, Switzerland)·2025
Same author

Two-Step Image Registration for Dual-Layer Flat-Panel Detectors.

Diagnostics (Basel, Switzerland)·2024
Same author

Self-Healing and Antifreezing/Antidrying Conductive Eutectohydrogel-Based Biosignal Monitoring Multisensors with Integrated Supercapacitor.

Small (Weinheim an der Bergstrasse, Germany)·2024
Same author

Measurements of the noise power spectrum for digital x-ray imaging devices.

Physics in medicine and biology·2023
Same author

Convex Combination of Images From Dual-Layer Detectors for High Detective Quantum Efficiencies.

IEEE transactions on bio-medical engineering·2023

Area of Science:

  • Digital Image Processing
  • Computer Vision

Background:

  • Blocking artifacts are a common distortion in Joint Photographic Experts Group (JPEG)-compressed images, degrading visual quality.
  • Existing postprocessing algorithms often involve iterative methods or complex filters, leading to high computational complexity.

Purpose of the Study:

  • To propose a new noniterative postprocessing algorithm for reducing blocking artifacts in JPEG-compressed images.
  • To improve both objective distortion metrics and subjective perceptual quality.
  • To achieve comparable or better results than existing methods with reduced computational complexity.

Main Methods:

  • A two-step approach involving low-pass filtering followed by prediction.
  • Predictors are constructed using a broken line regression model, generalized from projection onto quantization constraint sets.

Related Experiment Videos

  • Adaptive application of different predictors based on frequency components in the Discrete Cosine Transform (DCT) domain and local variance of DCT blocks.
  • Modification of the regression model for major-edge DCT blocks to address ringing artifacts.

    • Main Results:

    • The adaptive predictor strategy effectively handles smooth and detail regions, leading to appropriate blurring.
    • Demonstrated improvements in average distortion and perceptual quality compared to previous methods.
    • Successfully reduced ringing artifacts in major-edge DCT blocks through regression model modification.
    • The proposed algorithm achieves comparable or superior results with significantly less computational complexity.

    Conclusions:

    • The proposed noniterative algorithm effectively reduces blocking and ringing artifacts in JPEG images.
    • Adaptive prediction based on DCT frequency components and local variance is key to performance.
    • The algorithm offers a computationally efficient alternative to existing postprocessing techniques.