Optimized image steganography using heterogeneous image embedding techniques
Journal of Discrete Mathematical Sciences & Cryptography · 2026 · 29 (6) · 2431-2440
DOI: 10.47974/jdmsc-2751 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Discrete Mathematical Sciences & Cryptography, 2026. Verfasst von Shruti Jairath und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The need of safe information hiding has been on the rise in the context of the digital communication with the growth of sharing of online data spread exponentially. This paper presents a new method called the optimized Heterogeneous Image Steganography Network (HISN) in which the various attributes of heterogeneous images are used to greatly improve data hiding and security. HISN makes secret images highly invisible and accurate under cover images using convolutional neural networks. HISN, being made of three networks, which are interconnected, the Preparation Network, the Hiding Network and the Reveal Network is highly sensitive to control the optimized embedding processes and the decoding processes. HISN is strong and reliable, with a training accuracy of 92.64% and a testing accuracy of 90.47. The cover images are similar to the original images and the results of the encoder images, the secret images are similar to the original images, so there is an effective data hiding and recovery.
Zitieren
APA
Jairath, S., & Pal, K. (2026). Optimized image steganography using heterogeneous image embedding techniques. Journal of Discrete Mathematical Sciences & Cryptography, 29(6), 2431-2440. https://doi.org/10.47974/jdmsc-2751
Vancouver
Jairath S, Pal K. Optimized image steganography using heterogeneous image embedding techniques. Journal of Discrete Mathematical Sciences & Cryptography. 2026;29(6):2431-2440. doi:10.47974/jdmsc-2751
BibTeX
@article{Jairath2026,
author = {Jairath, Shruti and Pal, Kunwar},
title = {Optimized image steganography using heterogeneous image embedding techniques},
journal = {Journal of Discrete Mathematical Sciences & Cryptography},
year = {2026},
volume = {29},
number = {6},
pages = {2431--2440},
doi = {10.47974/jdmsc-2751}
}
EUVIMED