Abstract:Adversarial steganography aims to fool steganalysis models and enhance embedding security. Its appli-cation in the sequential min-max steganographic game has recently shown state-of-the-art embedding security. However, current methods suffer from high computational costs and inferior convergence. These issues could be caused by unnecessary optimization on distortions and the monotonous criterion for ad-versaries. This paper introduces a novel and concise scheme for conducting adversarial steganography and refines the criterion for adversaries. The suggested adversarial optimizations leave out distortions and focus on optimizing the stego distribution, whose difference with the intrinsic cover distribution is the primary factor affecting embedding security. These optimization techniques are accomplished by utilizing gradient descent on embedding modification probabilities and introducing a novel protocol for cover enhancement. Moreover, the proposed criteria for adversaries prioritize enhancing the similarity between each cover-stego pair, rather than evading steganalysis models, to break monotony and enhance convergence. The discussion and refinement of the loss function and termination setting of the optimiza-tion process are also provided. The proposed method demonstrates its benefits in increased security and computation efficiency in experiments. After collaborating with the steganographic min-max game, the state-of-the-art embedding security and stronger resistance to various and unexpected steganalysis mod-els are seen. & COPY; 2023 Elsevier B.V. All rights reserved.

Joint Cover-Selection and Payload-Allocation by Steganographic Distortion Optimization.

A New Steganographic Method Resisting Statistical Attack

Secure Steganography Based on Binary Particle Swarm Optimization

Practical Cover Selection for Steganography

Decomposing Joint Distortion for Adaptive Steganography

Diverse Batch Steganography Using Model-based Selection and Double-layered Payload Assignment

Secure Cover Selection for Steganography

Evolving Distortion Function by Exploiting the Differences among Comparable Adaptive Steganography

Steganography with Histogram-Preserving and Distortion-Constraining Properties

Robust steganography resisting JPEG compression by improving selection of cover element

Video Steganography Based on Modification Probability Transformation and Non-additive Embedding Distortion

Ensemble Stego Selection for Enhancing Image Steganography

Adaptive JPEG steganography with new distortion function

Motion Vector Modification Distortion Analysis-Based Payload Allocation for Video Steganography

Asymmetric Distortion Function for JPEG Steganography Using Block Artifact Compensation.

Hybrid Distortion Function for JPEG Steganography

Visually Optimized High-Quality Jpeg Steganography

On Improving Distortion Functions for JPEG Steganography.

Enhancing image steganography via adversarial optimization of the stego distribution

Steganographic Distortion Function for Enhanced Images.

Distortion Function for JPEG Steganography Based on Image Texture and Correlation in DCT Domain