An Effective Linguistic Steganalysis Framework Based on Hierarchical Mutual Learning

Yiming Xue,Lingzhi Kong,Wanli Peng,Ping Zhong,Juan Wen

DOI: https://doi.org/10.1016/j.ins.2021.11.086

IF: 8.1

2022-01-01

Information Sciences

Abstract:In recent years, the study of linguistic steganalysis has been focused on altering network structure, such as replacing the basic neural units or increasing the model depth, which inevitably increases computational overhead and restricts further improvement in resource-constrained scenarios. In this paper, instead of relying on complex neural networks, we propose an alternative linguistic steganalysis framework based on hierarchical mutual learning to achieve higher detection accuracy with less inference time and model size. In the proposed method, networks with either identical or different structures are trained cooperatively to learn distinct text features from each other. To this end, in addition to the supervised learning loss function, we construct three mimicry loss functions at different feature extraction stages, which can integrate steganalytic features from various levels. Finally, we illustrate how the steganalysis framework can be extended from two networks to multiple networks. Four representative steganalysis networks with different structures are employed to verify the effectiveness of our framework. The experimental results show that the proposed framework can effectively assist networks with fewer resources to perform better in model size, inference time, and detection accuracy than state-of-the-art steganalysis algorithms.