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 Video

Updated: Jan 9, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

996

Secure edge-guided adaptive image steganography using HED-based attention maps and CNN.

Rana Alrawashdeh1, Sultan Almuhammadi1,2, Mahmood Niazi1,2

  • 1Department of Information and Computer Science, College of Computing and Mathematics, King Fahd University of Petroleum and Minerals, 31261, Dhahran, Saudi Arabia.

Scientific Reports
|December 2, 2025
PubMed
Summary

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

In situ vaccination, low-dose targeted radionuclide therapy, and immune checkpoint inhibition eradicate poorly immunogenic metastatic tumors in murine cancer models.

Cancer immunology research·2026
Same author

Rhizospheric Microbes and Nanoparticles Synergize to Enhance Plant Immune Responses.

Plant-environment interactions (Hoboken, N.J.)·2026
Same author

Optimized steganographic embedding guided by snake algorithm and fusion-aware attention maps.

Scientific reports·2026
Same author

Influence of internal migration on antenatal care utilization in Bangladesh: Findings from a nationally representative cross-sectional survey.

PloS one·2026
Same author

Revolutionizing Transcriptomics: From Single-Cell Insights to RNA-based Interventions.

SLAS technology·2026
Same author

Transportation welfare under commercial conditions: A multifactorial assessment of stress physiology, meat quality, and microbial load in broiler chickens.

Poultry science·2026

This study introduces a novel image steganography system using deep learning and edge detection for secure data embedding. The method balances capacity and distortion, achieving high imperceptibility and security against steganalysis.

Area of Science:

  • Computer Science
  • Information Security
  • Digital Image Processing

Background:

  • Image steganography enables covert data transmission by hiding information within cover images.
  • Existing methods often struggle to balance embedding capacity, image distortion, and security.
  • Deep learning offers potential for more sophisticated and adaptive steganographic techniques.

Purpose of the Study:

  • To propose an advanced steganographic system utilizing edge-aware attention mechanisms within deep learning frameworks.
  • To enhance adaptive data embedding by directing operations based on image features.
  • To optimize the embedding process for improved capacity, imperceptibility, and security.

Main Methods:

  • Employing Holistically-Nested Edge Detection (HED) to extract edge maps and convert them into attention maps.
Keywords:
BlowfishFused MapsParticle Swarm Optimization (PSO)Steganography

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

725

Related Experiment Videos

Last Updated: Jan 9, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

996
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

725
  • Utilizing attention maps to guide adaptive bit embedding in cover images, adjusting bits per pixel based on attention strength.
  • Implementing a custom adaptive Least Significant Bits (LSBs) strategy guided by a trained encoder-decoder Convolutional Neural Network (CNN).
  • Optimizing the embedding rules using a genetic algorithm (GA) to fine-tune attention map thresholds.

    • Main Results:

    • The system achieves high Peak Signal to Noise Ratio (PSNR) values (e.g., 60.72-61.20 dB at 0.1 bits per pixel) and Structural Similarity Index (SSIM) values (0.9995-0.9996).
    • Steganalysis techniques (Xu-Net, Ye-Net, RS analysis) failed to reliably detect hidden data, indicating strong security.
    • The method demonstrates robustness against salt-and-pepper noise and maintains reasonable performance against other common image attacks.

    Conclusions:

    • The proposed deep learning-based image steganography system effectively balances embedding capacity, imperceptibility, and security.
    • Edge-aware attention mechanisms significantly improve adaptive data embedding.
    • The system offers a low computational cost solution with promising resistance to various attacks.