Zero-Shot Interpretable Image Steganalysis for Invertible Image Hiding

TL;DR

This paper introduces a zero-shot, interpretable image steganalysis framework that detects and recovers hidden information in images, improving generalization across datasets and architectures for invertible image hiding methods.

Contribution

It presents a unified framework integrating image hiding, revealing, and steganalysis, with a residual augmentation strategy to enhance cross-scenario robustness.

Findings

Outperforms existing steganalysis techniques on benchmark datasets.

Effectively recovers secret information embedded in stego images.

Generalizes well across different datasets and model architectures.

Abstract

Image steganalysis, which aims at detecting secret information concealed within images, has become a critical countermeasure for assessing the security of steganography methods, especially the emerging invertible image hiding approaches. However, prior studies merely classify input images into two categories (i.e., stego or cover) and typically conduct steganalysis under the constraint that training and testing data must follow similar distribution, thereby hindering their application in real-world scenarios. To overcome these shortcomings, we propose a novel interpretable image steganalysis framework tailored for invertible image hiding schemes under a challenging zero-shot setting. Specifically, we integrate image hiding, revealing, and steganalysis into a unified framework, endowing the steganalysis component with the capability to recover the secret information embedded in stego images. Additionally, we elaborate a simple yet effective residual augmentation strategy for generating stego images to further enhance the generalizability of the steganalyzer in cross-dataset and cross-architecture scenarios. Extensive experiments on benchmark datasets demonstrate that our proposed approach significantly outperforms the existing steganalysis techniques for invertible image hiding schemes.