GAN-based carrier image synthesis steganography method

Abstract

The invention relates to a GAN-based carrier image synthesis steganography method, which comprises the steps of cutting each image in a data set into images with the same size, and forming a real image data set; constructing a generation network G, a discrimination network D, an evidence obtaining network F and an extraction network E, initializing parameters in the generation network G, the discrimination network D and the extraction network E, and setting the parameters in the evidence obtaining network F as preset values; training the initialized generation network G, discrimination networkD and extraction network E by using a batch training mode to obtain a trained generation network G, discrimination network D and extraction network E; inputting secret information to be embedded anda preset secret key into the trained generation network G to obtain a synthetic carrier image; and inputting the synthesized carrier image and the preset secret key into the trained extraction networkE, and extracting secret information. According to the method, the security of the algorithm is enhanced, and the naturalness of the synthesized carrier image is improved.

Description

technical field [0001] The invention relates to the field of image steganography, in particular to a GAN-based steganographic method for carrier image synthesis. Background technique [0002] Steganography is an important branch of information hiding technology, and its goal is to hide secret information in a digital carrier (such as images, audio and video, etc.) in an imperceptible way, and send it to the receiver. Modern steganographic methods are mainly divided into three basic architectures: carrier modification, carrier selection, and carrier synthesis. Among the above three basic architectures, the development of carrier selection and carrier synthesis steganography lags behind compared to carrier modification steganography. Among them, the carrier synthesis steganography method can only generate non-semantic images such as textures and fingerprints, which leads to very limited use scenarios. [0003] In recent years, among various deep neural networks, GAN has rece...

AI Technical Summary

Problems solved by technology

Therefore, although the method proposed in the above literature alleviates the problem of "unable to generate semantic carrier image" to a certain extent, the naturalness of the synthesized carrier image still needs to be improved.

Secondly, there is no key involved in this method, so the security of secret information cannot be protected when the method is made public, that is, the Kirkhovsky principle in cryptography is not satisfied

Finally, synthetic cover images are vulnerable to synthetic image forensic detection, thereby exposing the inspector

Images