Accession Number:

ADA616034

Title:

Universal Batch Steganalysis

Descriptive Note:

Final rept. 1 Jul 2011-30 Jun 2014

Corporate Author:

CZECH TECHNICAL UNIV PRAGUE

Personal Author(s):

• Pevny, Tomas
• Ker, Andrew D.

Report Date:

2014-06-01

Pagination or Media Count:

18.0

Abstract:

Steganography is the hiding of information within an innocent cover such a technology is becoming increasingly desirable for those who wish to evade network monitoring or to disguise their circle of correspondents. On the other hand, it presents a challenge for those with a legitimate need to monitor networks. The aim of steganalysis is to decide whether internet traffic contains hidden data, and the literature has developed rapidly in the last ten years. For full literature surveys, see the papers 1-8. However, it has developed in a particularly specialised direction, to binary classification innocent cover vs. guilty stego of single objects using machine learning methods, applied almost exclusively to grayscale images. The practical application of such a detector is uncertain, as it requires potentially unrealistic knowledge by the detector the exact embedding method used, usually the size of the embedded payload, and training data from the same source as that being examined. Furthermore, it does not generalise well to detection of multiple objects.

Descriptors:

• *STEGANOGRAPHY
• COMMUNICATIONS NETWORKS
• CZECH REPUBLIC
• LEARNING MACHINES
• MONITORING
• NANOPARTICLES

Subject Categories:

• Cybernetics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE