Can Machine Learn Steganography? - Implementing LSB Substitution and Matrix Coding Steganography with Feed-Forward Neural Networks

In recent years, due to the powerful abilities to deal with highly complex tasks, the artificial neural networks (ANNs) have been studied in the hope of achieving human-like performance in many applications. Since the ANNs have the ability to approximate complex functions from observations, it is st...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Han-Zhou, Wu, Hong-Xia, Wang, Yun-Qing, Shi
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 16.06.2016
Subjects
Artificial neural networks
Coding
Human performance
Neural networks
Steganography
Substitutes
Task complexity
Online AccessGet full text

Cover

More Information
Summary:In recent years, due to the powerful abilities to deal with highly complex tasks, the artificial neural networks (ANNs) have been studied in the hope of achieving human-like performance in many applications. Since the ANNs have the ability to approximate complex functions from observations, it is straightforward to consider the ANNs for steganography. In this paper, we aim to implement the well-known LSB substitution and matrix coding steganography with the feed-forward neural networks (FNNs). Our experimental results have shown that, the used FNNs can achieve the data embedding operation of the LSB substitution and matrix coding steganography. For steganography with the ANNs, though there may be some challenges to us, it would be very promising and valuable to pay attention to the ANNs for steganography, which may be a new direction for steganography.
ISSN:2331-8422