Translator Disclaimer
10 February 2011 A new paradigm for steganalysis via clustering
Author Affiliations +
Proceedings Volume 7880, Media Watermarking, Security, and Forensics III; 78800U (2011)
Event: IS&T/SPIE Electronic Imaging, 2011, San Francisco Airport, California, United States
We propose a new paradigm for blind, universal, steganalysis in the case when multiple actors transmit multiple objects, with guilty actors including some stego objects in their transmissions. The method is based on clustering rather than classification, and it is the actors which are clustered rather than their individual transmitted objects. This removes the need for training a classifier, and the danger of training model mismatch. It effectively judges the behaviour of actors by assuming that most of them are innocent: after performing agglomerative hierarchical clustering, the guilty actor(s) are clustered separately from the innocent majority. A case study shows that this works in the case of JPEG images. Although it is less sensitive than steganalysis based on specifically-trained classifiers, it requires no training, no knowledge of the embedding algorithm, and attacks the pooled steganalysis problem.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Andrew D. Ker and Tomás Pevný "A new paradigm for steganalysis via clustering", Proc. SPIE 7880, Media Watermarking, Security, and Forensics III, 78800U (10 February 2011);

Back to Top