Secure transmission of ocean images using deep learning‐based data hiding
Abstract Data hiding has become a hot research topic in recent years due to increased attention placed on the copyright protection of ocean images and related digital records. Further, high image volumes put enormous pressure on transmission bandwidth and storage capabilities. This paper proposes an...
Saved in:
Published in | Expert systems |
---|---|
Main Authors | , , |
Format | Journal Article |
Language | English |
Published |
11.10.2023
|
Online Access | Get full text |
Cover
Loading…
Summary: | Abstract
Data hiding has become a hot research topic in recent years due to increased attention placed on the copyright protection of ocean images and related digital records. Further, high image volumes put enormous pressure on transmission bandwidth and storage capabilities. This paper proposes an innovative deep learning‐based data‐hiding technique for ocean images. First, a down‐sampling scheme is applied to compress the secret mark before embedding it in the host media. Then, a convolutional neural network is used to embed and recover compressed marks into or from the host ocean image. Finally, a generative adversarial network‐based reconstruction network is used to reconstruct the high‐quality mark image. Our experiments show that the proposed work not only maintains high imperceptibility and robustness against many attacks but also provides better data‐hiding performance than related works. |
---|---|
ISSN: | 0266-4720 1468-0394 |
DOI: | 10.1111/exsy.13469 |