Formalizing Cover-source Mismatch as a Robust Optimization
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00364285" target="_blank" >RIV/68407700:21230/22:00364285 - isvavai.cz</a>
Result on the web
<a href="https://scholar.google.com/citations?view_op=view_citation&hl=en&user=MnXqDssAAAAJ&sortby=pubdate&citation_for_view=MnXqDssAAAAJ:Z5m8FVwuT1cC" target="_blank" >https://scholar.google.com/citations?view_op=view_citation&hl=en&user=MnXqDssAAAAJ&sortby=pubdate&citation_for_view=MnXqDssAAAAJ:Z5m8FVwuT1cC</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Formalizing Cover-source Mismatch as a Robust Optimization
Original language description
Cover-source mismatch (CSM) refers to the use of a steganographic detector on images with a very different probability distribution it has been trained on. This can have a detrimental effect on its accuracy preventing the use of modern steganalytic tools outside laboratories. Despite CSM being introduced almost fifteen years ago, there is no formal definition and no adopted measures for comparing different solutions. This work, therefore, formalizes the cover-source mismatch and proposes and discusses possible error measures. Equipped with these tools, we propose a principled approach to train holistic detectors while minimizing the effects of CSM and experimentally compare them to the prior art, discussing their strength and weaknesses.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of European Signal Processing Conference
ISBN
978-1-6654-6798-8
ISSN
—
e-ISSN
—
Number of pages
5
Pages from-to
1042-1046
Publisher name
EUSIPCO
Place of publication
Belgrade
Event location
Belgrade
Event date
Aug 29, 2022
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—