Unsupervised steganalysis based on artificial training sets

In this paper, an unsupervised steganalysis method that combines artificial training sets and supervised classification is proposed. We provide a formal framework for unsupervised classification of stego and cover images in the typical situation of targeted steganalysis (i.e., for a known algorithm...

ver descrição completa

Detalhes bibliográficos
Autores: Lerch-Hostalot, Daniel, Megias, David
Formato: artículo
Estado:Versión enviada para evaluación y publicación
Fecha de publicación:2015
País:España
Recursos:Universitat Oberta de Catalunya (UOC)
Repositorio:O2, repositorio institucional de la UOC
OAI Identifier:oai:openaccess.uoc.edu:10609/82325
Acesso em linha:http://hdl.handle.net/10609/82325
Access Level:acceso abierto
Palavra-chave:unsupervised steganalysis
cover source mismatch
machine learning
esteganàlisi no supervisat
desajustament de la font de portada
aprenentatge automàtic
esteganálisis no supervisado
desajuste de la fuente de portada
aprendizaje automático
Artificial intelligence -- Engineering applications
Intel·ligència artificial -- Aplicacions a l'enginyeria
Inteligencia artificial -- Aplicaciones a la ingeniería
id ES_bd1647da8aac0645427f9b36845f25bb
oai_identifier_str oai:openaccess.uoc.edu:10609/82325
network_acronym_str ES
network_name_str España
repository_id_str
spelling Unsupervised steganalysis based on artificial training setsLerch-Hostalot, DanielMegias, Davidunsupervised steganalysiscover source mismatchmachine learningesteganàlisi no supervisatdesajustament de la font de portadaaprenentatge automàticesteganálisis no supervisadodesajuste de la fuente de portadaaprendizaje automáticoArtificial intelligence -- Engineering applicationsIntel·ligència artificial -- Aplicacions a l'enginyeriaInteligencia artificial -- Aplicaciones a la ingenieríaIn this paper, an unsupervised steganalysis method that combines artificial training sets and supervised classification is proposed. We provide a formal framework for unsupervised classification of stego and cover images in the typical situation of targeted steganalysis (i.e., for a known algorithm and approximate embedding bit rate). We also present a complete set of experiments using (1) eight different image databases, (2) image features based on Rich Models, and (3) three different embedding algorithms: Least Significant Bit (LSB) matching, Highly undetectable steganography (HUGO) and Wavelet Obtained Weights (WOW). We show that the experimental results outperform previous methods based on Rich Models in the majority of the tested cases. At the same time, the proposed approach bypasses the problem of Cover Source Mismatch -when the embedding algorithm and bit rate are known- since it removes the need of a training database when we have a large enough testing set. Furthermore, we provide a generic proof of the proposed framework in the machine learning context. Hence, the results of this paper could be extended to other classification problems similar to steganalysis.Engineering Applications of Artificial IntelligenceUniversitat Oberta de Catalunya. Internet Interdisciplinary Institute (IN3)201820182015info:eu-repo/semantics/articleinfo:eu-repo/semantics/submittedVersionapplication/pdfhttp://hdl.handle.net/10609/82325reponame:O2, repositorio institucional de la UOCinstname:Universitat Oberta de Catalunya (UOC)InglésEngineering Applications of Artificial Intelligence, 2016, 50https://doi.org/10.1016/j.engappai.2015.12.013CC BY-NC-NDhttp://creativecommons.org/licenses/by-nc-nd/3.0/es/info:eu-repo/semantics/openAccessoai:openaccess.uoc.edu:10609/823252026-05-28T12:42:01Z
dc.title.none.fl_str_mv Unsupervised steganalysis based on artificial training sets
title Unsupervised steganalysis based on artificial training sets
spellingShingle Unsupervised steganalysis based on artificial training sets
Lerch-Hostalot, Daniel
unsupervised steganalysis
cover source mismatch
machine learning
esteganàlisi no supervisat
desajustament de la font de portada
aprenentatge automàtic
esteganálisis no supervisado
desajuste de la fuente de portada
aprendizaje automático
Artificial intelligence -- Engineering applications
Intel·ligència artificial -- Aplicacions a l'enginyeria
Inteligencia artificial -- Aplicaciones a la ingeniería
title_short Unsupervised steganalysis based on artificial training sets
title_full Unsupervised steganalysis based on artificial training sets
title_fullStr Unsupervised steganalysis based on artificial training sets
title_full_unstemmed Unsupervised steganalysis based on artificial training sets
title_sort Unsupervised steganalysis based on artificial training sets
dc.creator.none.fl_str_mv Lerch-Hostalot, Daniel
Megias, David
author Lerch-Hostalot, Daniel
author_facet Lerch-Hostalot, Daniel
Megias, David
author_role author
author2 Megias, David
author2_role author
dc.contributor.none.fl_str_mv Universitat Oberta de Catalunya. Internet Interdisciplinary Institute (IN3)
dc.subject.none.fl_str_mv unsupervised steganalysis
cover source mismatch
machine learning
esteganàlisi no supervisat
desajustament de la font de portada
aprenentatge automàtic
esteganálisis no supervisado
desajuste de la fuente de portada
aprendizaje automático
Artificial intelligence -- Engineering applications
Intel·ligència artificial -- Aplicacions a l'enginyeria
Inteligencia artificial -- Aplicaciones a la ingeniería
topic unsupervised steganalysis
cover source mismatch
machine learning
esteganàlisi no supervisat
desajustament de la font de portada
aprenentatge automàtic
esteganálisis no supervisado
desajuste de la fuente de portada
aprendizaje automático
Artificial intelligence -- Engineering applications
Intel·ligència artificial -- Aplicacions a l'enginyeria
Inteligencia artificial -- Aplicaciones a la ingeniería
description In this paper, an unsupervised steganalysis method that combines artificial training sets and supervised classification is proposed. We provide a formal framework for unsupervised classification of stego and cover images in the typical situation of targeted steganalysis (i.e., for a known algorithm and approximate embedding bit rate). We also present a complete set of experiments using (1) eight different image databases, (2) image features based on Rich Models, and (3) three different embedding algorithms: Least Significant Bit (LSB) matching, Highly undetectable steganography (HUGO) and Wavelet Obtained Weights (WOW). We show that the experimental results outperform previous methods based on Rich Models in the majority of the tested cases. At the same time, the proposed approach bypasses the problem of Cover Source Mismatch -when the embedding algorithm and bit rate are known- since it removes the need of a training database when we have a large enough testing set. Furthermore, we provide a generic proof of the proposed framework in the machine learning context. Hence, the results of this paper could be extended to other classification problems similar to steganalysis.
publishDate 2015
dc.date.none.fl_str_mv 2015
2018
2018
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/submittedVersion
format article
status_str submittedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10609/82325
url http://hdl.handle.net/10609/82325
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Engineering Applications of Artificial Intelligence, 2016, 50
https://doi.org/10.1016/j.engappai.2015.12.013
dc.rights.none.fl_str_mv CC BY-NC-ND
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv CC BY-NC-ND
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Engineering Applications of Artificial Intelligence
publisher.none.fl_str_mv Engineering Applications of Artificial Intelligence
dc.source.none.fl_str_mv reponame:O2, repositorio institucional de la UOC
instname:Universitat Oberta de Catalunya (UOC)
instname_str Universitat Oberta de Catalunya (UOC)
reponame_str O2, repositorio institucional de la UOC
collection O2, repositorio institucional de la UOC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418170554515456
score 15.198674

Registos relacionados