Anzeige
Erschienen in:
2017 | OriginalPaper | Buchkapitel
Steganalysis Based on Statistical Properties of the Encrypted Messages
verfasst von : Valery Korzhik, Ivan Fedyanin, Artur Godlewski, Guillermo Morales-Luna
Erschienen in: Computer Network Security
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
It is introduced a new steganalytic method based on investigation of statistical properties of the extracted encrypted messages. It is assumed that the message is encrypted before embedding by some sufficiently strong cipher. The extracted messages are exposed to the testing on pseudorandomness, namely to NIST-tests. When passing these tests, the tested objects are assumed stego-objects, otherwise as cover objects. Support vector machine methodology is used to improve the testing process. Experiments demonstrate that the proposed algorithm is able to detect stegosystems as LSB-based embedding (both replacing and matching), with pseudorandom walks given by stego-keys, and with matrix embedding based on Hamming codes with reasonable reliability.