Steganographic Generative Adversarial Networks

TL;DR

This paper introduces a novel DCGAN-based model for generating image containers that enhance steganography security, effectively deceiving steganalysis tools and improving covert communication methods.

Contribution

The study presents a new generative model that produces more secure image containers for steganography, outperforming existing methods in evading steganalysis detection.

Findings

Successfully deceives steganalysis classifiers

Generates more secure image containers

Enhances steganographic embedding techniques

Abstract

Steganography is collection of methods to hide secret information ("payload") within non-secret information "container"). Its counterpart, Steganalysis, is the practice of determining if a message contains a hidden payload, and recovering it if possible. Presence of hidden payloads is typically detected by a binary classifier. In the present study, we propose a new model for generating image-like containers based on Deep Convolutional Generative Adversarial Networks (DCGAN). This approach allows to generate more setganalysis-secure message embedding using standard steganography algorithms. Experiment results demonstrate that the new model successfully deceives the steganography analyzer, and for this reason, can be used in steganographic applications.