Proceedings of the ACM Workshop on Information Hiding and Multimedia Security 2019
DOI: 10.1145/3335203.3335738
|View full text |Cite
|
Sign up to set email alerts
|

Detection of Classifier Inconsistencies in Image Steganalysis

Abstract: In this paper, a methodology to detect inconsistencies in classi cation-based image steganalysis is presented. e proposed approach uses two classi ers: the usual one, trained with a set formed by cover and stego images, and a second classi er trained with the set obtained a er embedding additional random messages into the original training set. When the decisions of these two classi ers are not consistent, we know that the prediction is not reliable. e number of inconsistencies in the predictions of a testing … Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
1
0
1

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(2 citation statements)
references
References 21 publications
0
1
0
1
Order By: Relevance
“…There exists work whose idea is similar to that of our work. In that work [12], its goal is to detect inconsistencies occurred during classification in image steganalysis by additionally embedding steganography. In other words, that work is a method that deal with the problem known as Cover Source Mismatch (CSM).…”
Section: Introductionmentioning
confidence: 99%
“…There exists work whose idea is similar to that of our work. In that work [12], its goal is to detect inconsistencies occurred during classification in image steganalysis by additionally embedding steganography. In other words, that work is a method that deal with the problem known as Cover Source Mismatch (CSM).…”
Section: Introductionmentioning
confidence: 99%
“…Их существенными недостатками являются: требовательность аналитических моделей, на которых основываются методы стегоанализа, к ряду параметров анализируемых контейнеров, что влияет на чувствительность моделей, и необходимость реализации полученных решений численными методами, что приводит к снижению оперативности и результативности процесса стегоанализа. В последнее время ряд исследований в области стегоанализа, в частности, цифровых изображений посвящен применению методов машинного обучения, основанных на применении искусственных нейронных сетей [1,3,4]. С использованием этих методов решается задача бинарной классификации цифровых изображений -разделения их множества на подмножества, содержащие и не содержащие стеговложения.…”
Section: Introductionunclassified