Improving Blind Steganalysis in Spatial Domain using a Criterion to Choose the Appropriate Steganalyzer between CNN and SRM+EC

TL;DR

This paper introduces a criterion to select the most effective steganalyzer, either CNN or SRM+EC, for a given image, improving blind detection performance across multiple spatial domain algorithms.

Contribution

It proposes a novel criterion for choosing between CNN and SRM+EC steganalyzers, enhancing detection accuracy in blind steganalysis of spatial domain images.

Findings

Detection error rates are reduced compared to individual methods.

The approach works effectively across different steganographic algorithms.

It demonstrates improved blind detection performance at various payloads.

Abstract

Conventional state-of-the-art image steganalysis approaches usually consist of a classifier trained with features provided by rich image models. As both features extraction and classification steps are perfectly embodied in the deep learning architecture called Convolutional Neural Network (CNN), different studies have tried to design a CNN-based steganalyzer. The network designed by Xu et al. is the first competitive CNN with the combination Spatial Rich Models (SRM) and Ensemble Classifier (EC) providing detection performances of the same order. In this work we propose a criterion to choose either the CNN or the SRM+EC method for a given input image. Our approach is studied with three different steganographic spatial domain algorithms: S-UNIWARD, MiPOD, and HILL, using the Tensorflow computing platform, and exhibits detection capabilities better than each method alone. Furthermore, as SRM+EC and the CNN are both only trained with a single embedding algorithm, namely MiPOD, the proposed method can be seen as an approach for blind steganalysis. In blind detection, error rates are respectively of 16% for S-UNIWARD, 16% for MiPOD, and 17% for HILL on the BOSSBase with a payload of 0.4 bpp. For 0.1 bpp, the respective corresponding error rates are of 39%, 38%, and 41%, and are always better than the ones provided by SRM+EC.