Detection of Classifier Inconsistencies in Image Steganalysis

TL;DR

This paper introduces a method to detect unreliable classifications in image steganalysis by comparing two classifiers trained under different conditions, helping identify when the classifier may be failing due to source mismatch or unseen steganographic methods.

Contribution

The paper proposes a novel inconsistency detection approach using dual classifiers to assess the reliability of steganalysis predictions, addressing issues like source mismatch and unknown steganography.

Findings

Inconsistencies correlate with classification errors.

The method predicts classifier reliability effectively.

Applicable to various steganalysis scenarios.

Abstract

In this paper, a methodology to detect inconsistencies in classification-based image steganalysis is presented. The proposed approach uses two classifiers: the usual one, trained with a set formed by cover and stego images, and a second classifier trained with the set obtained after embedding additional random messages into the original training set. When the decisions of these two classifiers are not consistent, we know that the prediction is not reliable. The number of inconsistencies in the predictions of a testing set may indicate that the classifier is not performing correctly in the testing scenario. This occurs, for example, in case of cover source mismatch, or when we are trying to detect a steganographic method that the classifier is no capable of modelling accurately. We also show how the number of inconsistencies can be used to predict the reliability of the classifier (classification errors).