Using contrastive learning to improve the performance of steganalysis schemes

TL;DR

This paper introduces a contrastive learning framework for steganalysis that enhances detection accuracy and generalization while reducing training time.

Contribution

It proposes the Steganalysis Contrastive Framework (SCF) and a novel Steganalysis Contrastive Loss (StegCL) to improve feature representation and efficiency.

Findings

Improves detection accuracy by up to 3%.

Enhances generalization of steganalysis models.

Reduces training time by 90%.

Abstract

To improve the detection accuracy and generalization of steganalysis, this paper proposes the Steganalysis Contrastive Framework (SCF) based on contrastive learning. The SCF improves the feature representation of steganalysis by maximizing the distance between features of samples of different categories and minimizing the distance between features of samples of the same category. To decrease the computing complexity of the contrastive loss in supervised learning, we design a novel Steganalysis Contrastive Loss (StegCL) based on the equivalence and transitivity of similarity. The StegCL eliminates the redundant computing in the existing contrastive loss. The experimental results show that the SCF improves the generalization and detection accuracy of existing steganalysis DNNs, and the maximum promotion is 2% and 3% respectively. Without decreasing the detection accuracy, the training time of using the StegCL is 10% of that of using the contrastive loss in supervised learning.