Publication #151

Detection of hiding in the least significant bit

O. Dabeer, K. Sullivan,U. Madhow, S. Chandrasekharan, B. S. Manjunath
Dept. of Electrical and Computer Engineering
University of California at Santa Barbara,
Santa Barbara, CA 93106
E-mail: onkar@ece.ucsb.edu

Abstract

We consider the problem of detecting hiding in the least significant bit (LSB) of images. Since the hiding rate is not known, this is a composite hypothesis testing problem. We show that under a mild condition on the host probability mass function (PMF), the optimal composite hypothesis testing problem is solved by a related optimal simple hypothesis testing problem. We then develop practical tests based on the optimal test and exhibit their superiority over Stegdetect, a popular steganalysis method used in practice.

[PDF] [BibTex]

O. Dabeer, K. Sullivan, U. Madow, S. Chandrasekaran and B. S. Manjunath,
"Detection of hiding in the least significant bit"
Proc. Conference on Information Sciences and Systems, Mar. 2003.

DB ID: 151, VRL ID: 113, Lab: VRL, Target: Proceedings
Subject: [Watermarking_Hiding] « Look-up more
Project: [Steganalysis] « Look-up more

» More: Advanced search