Invisible Steganography via Generative Adversarial Networks

TL;DR

This paper introduces ISGAN, a novel CNN-based steganography method that enhances invisibility, security, and image quality by hiding secrets in the Y channel and using GANs with a mixed loss function, achieving state-of-the-art results.

Contribution

The paper presents a new CNN architecture, ISGAN, that improves invisibility, security, and image realism in steganography by leveraging GANs and a specialized loss function.

Findings

Achieves state-of-the-art performance on multiple datasets

Enhances invisibility by hiding secrets only in the Y channel

Strengthens security through GAN-based divergence minimization

Abstract

Nowadays, there are plenty of works introducing convolutional neural networks (CNNs) to the steganalysis and exceeding conventional steganalysis algorithms. These works have shown the improving potential of deep learning in information hiding domain. There are also several works based on deep learning to do image steganography, but these works still have problems in capacity, invisibility and security. In this paper, we propose a novel CNN architecture named as \isgan to conceal a secret gray image into a color cover image on the sender side and exactly extract the secret image out on the receiver side. There are three contributions in our work: (i) we improve the invisibility by hiding the secret image only in the Y channel of the cover image; (ii) We introduce the generative adversarial networks to strengthen the security by minimizing the divergence between the empirical probability distributions of stego images and natural images. (iii) In order to associate with the human visual system better, we construct a mixed loss function which is more appropriate for steganography to generate more realistic stego images and reveal out more better secret images. Experiment results show that ISGAN can achieve start-of-art performances on LFW, Pascal VOC2012 and ImageNet datasets.