In these applications, one of the most important aspects is to hide information in a cover image whithout suffering any alteration. Nowadays, image steganography has an important role in hiding information in advanced applications, such as medical image communication, confidential communication and secret data storing, protection of data alteration, access control system for digital content distribution and media database systems. Moreover, we show the flexibility of the proposed method in terms of hiding multiple images for different receivers and obfuscating the secret image. We demonstrate the feasibility of our SinGAN approach in terms of extraction accuracy and model security. The stego SinGAN, behaving as the original SinGAN, is publicly communicated only the receiver with the embedding key is able to extract the secret image.
#ANOTHER NAME FOR OUTGUESS PATCH#
We hide the secret image by fitting a deterministic mapping from a fixed set of noise maps (generated by an embedding key) to the secret image during patch distribution learning. As an instantiation, we adopt a SinGAN, a pyramid of generative adversarial networks (GANs), to learn the patch distribution of one cover image. Specifically, we use a DNN to model the probability density of cover images, and hide a secret image in one particular location of the learned distribution. In this work, we describe a different computational framework to hide images in deep probabilistic models. This scheme may suffer from several limitations regarding practicability, security, and embedding capacity.
A prevailing scheme is to train an autoencoder, consisting of an encoding network to embed (or transform) secret messages in (or into) a carrier, and a decoding network to extract the hidden messages. Data hiding with deep neural networks (DNNs) has experienced impressive successes in recent years.
0 kommentar(er)
