Dig a Hole and Fill in Sand: Adversary and Hiding Decoupled Steganography

Weixuan Tang, Haoyu Yang, Yuan Rao, Zhili Zhou, Fei Peng

Abstract:

Deep steganography is a technique that imperceptibly hides secret information into image by neural networks. Existing networks consist of two components, including a hiding component for information hiding and an adversary component for countering against steganalyzers.
However, these two components are two ends of the seesaw, and it is difficult to balance the tradeoff between message extraction accuracy and security performance by joint optimization. To address the issues, this paper proposes a steganographic method called AHDeS (Adversary-Hiding-Decoupled Steganography) under the Dig-and-Fill paradigm, wherein the adversary and hiding components can be decoupled into an optimization-based adversary module in the digging process and an INN-based hiding network in the filling process. Specfically in the training stage, the INN is first trained for acquiring the ability of message embedding. In the deployment stage, given the well-trained and fixed INN, the cover image is first iteratively optimized for enhancing the security performance against steganalyzers, followed by the actual message embedding by the INN. Owing to the reversibility of the INN, security performance can be enhanced without sacrificing message extraction accuracy. Experimental results show that AHDeS can achieve the state-of-the-art security performance and visual quality while maintaining satisfied message extraction accuracy.

Primary Subject Area: [Experience] Multimedia Applications

Relevance To Conference: Multimedia steganography is a technique of secret communication, which imperceptibly hides secret information into multimedia carriers. However, modifying the carriers for message embedding would introduce artifacts, which may be detected by the steganalyzers. Therefore, implementing information hiding while obtaining high security performance is the key issue in steganography. Existing deep steganographic methods consist of two main components, including a hiding component for information hiding and an adversary component for countering against steganalyzers. However, jointly training these two components would encounter the difficulties in balancing the tradeoff between message extraction accuracy and security performance. This paper proposes a brand new paradigm called Dig-and-Fill, and proposes a steganographic method called AHDeS, wherein the adversary and hiding components can be decoupled into an optimization-based adversary module in the digging process and an INN-based hiding network in the filling process. In this manner, security performance can be enhanced without sacrificing message extraction accuracy. The decoupling characteristics in the Dig-and-Fill paradigm is general for different multimedia carriers. In this paper, we first apply it to image steganography, and would further extend it to other multimedia such as audio, video, and so on.

Submission Number: 3421