2026-04-13T07:36:22Zhttps://recercat.cat/oai/request

oai:recercat.cat:2072/3622102024-12-13T09:30:45Zcom_2072_228903com_2072_96873com_2072_96852com_2072_2015col_2072_228904

00925njm 22002777a 4500 dc Lerch Hostalot, Daniel author Megías Jiménez, David author 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). Lerch-Hostalot, D. & Megías, D. (2019). Detection of classifier inconsistencies in image steganalysis. 7th ACM Workshop on Information Hiding and Multimedia Security. Proceedings, 2019 (), 222-229. doi: 10.1145/3335203.3335738 9781450368216 10.1145/3335203.3335738 http://hdl.handle.net/10609/100966 steganalysis cover source mismatch machine learning estegoanálisis aprendizaje automático desajuste de la fuente de portada estegoanàlisi aprenentatge automàtic desajustament de la font de portada Computer security Seguretat informàtica Seguridad informática Detection of classifier inconsistencies in image steganalysis