The Cover Source Mismatch Problem in Deep-Learning Steganalysis - Université de technologie de Troyes Access content directly
Conference Papers Year :

The Cover Source Mismatch Problem in Deep-Learning Steganalysis


This paper studies the problem of Cover Source Mismatch (CSM) in steganalysis, i.e. the impact of a testing set which does not originate from the same source than the training set. In this study, the trained steganalyzer uses state of the art deep-learning architecture prone to better generalization than feature-based steganalysis. Different sources such as the sensor model, the ISO sensitivity, the processing pipeline and the content, are investigated. Our conclusions are that, on one hand, deep learning steganalysis is still very sensitive to the CSM, on the other hand, the holistic strategy leverages the good generalization properties of deep learning to reduce the CSM with a relatively small number of training samples.
Fichier principal
Vignette du fichier
Giboulot_EUSIPCO_2022.pdf (298.76 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03694662 , version 1 (13-06-2022)


  • HAL Id : hal-03694662 , version 1


Quentin Giboulot, Patrick Bas, Rémi Cogranne, Dirk Borghys. The Cover Source Mismatch Problem in Deep-Learning Steganalysis. European Signal Processing Conference, Aug 2022, Belgrade, Serbia. ⟨hal-03694662⟩
112 View
118 Download


Gmail Facebook Twitter LinkedIn More