A Robust Coverless Image Steganography Method Based on Face Recognition and Camouflage Image

A Robust Coverless Image Steganography Method Based on Face Recognition and Camouflage Image

연구 분야: Strategies

학회: 2024 4th Asia Conference on Information Engineering (ACIE)

Information exchange and communication have been gradually integrated into life due to the development of technology. With the awareness of information security rising, image steganography which uses images as a carrier is also gradually growing in the field of data hiding. Compared with traditional embedded steganography, coverless image steganography has been attracting much attention in recent years. Coverless image steganography makes it possible to hide secret information without modifying the image information. However, the existing coverless image steganography still has some drawbacks, such as the huge size of auxiliary information, the inability to resist geometric attacks, and low image-visual correlation. To resolve the above shortcoming, we propose a coverless image steganography method based on face recognition and camouflage image. We use a camouflage image instead of the stego-image for transmission. To enhance robustness, we also propose a retrieval algorithm based on Structural Similarity Index to obtain images that have not been attacked. Further, we reduce the time complexity of our method through face recognition. In experiments, we utilized two datasets CACD and CelebA. The experimental results show that our method is highly robust and has behavioral security without the auxiliary information.