Digital Cardan Grille: A Modern Approach for Information Hiding
This is an incremental improvement for steganography applications, offering a modern approach to hiding information in images.
The paper tackles the problem of information hiding by proposing a Digital Cardan Grille framework that uses semantic image inpainting and GANs to embed secret messages into corrupted image regions, with experimental results confirming its feasibility.
In this paper, a new framework for construction of Cardan grille for information hiding is proposed. Based on the semantic image inpainting technique, the stego image are driven by secret messages directly. A mask called Digital Cardan Grille (DCG) for determining the hidden location is introduced to hide the message. The message is written to the corrupted region that needs to be filled in the corrupted image in advance. Then the corrupted image with secret message is feeded into a Generative Adversarial Network (GAN) for semantic completion. The adversarial game not only reconstruct the corrupted image , but also generate a stego image which contains the logic rationality of image content. The experimental results verify the feasibility of the proposed method.