Share Email Print

Proceedings Paper

Quantitative steganalysis using rich models
Author(s): Jan Kodovský; Jessica Fridrich
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In this paper, we propose a regression framework for steganalysis of digital images that utilizes the recently proposed rich models – high-dimensional statistical image descriptors that have been shown to substantially improve classical (binary) steganalysis. Our proposed system is based on gradient boosting and utilizes a steganalysis-specific variant of regression trees as base learners. The conducted experiments confirm that the proposed system outperforms prior quantitative steganalysis (both structural and feature-based) across a wide range of steganographic schemes: HUGO, LSB replacement, nsF5, BCHopt, and MME3.

Paper Details

Date Published: 22 March 2013
PDF: 11 pages
Proc. SPIE 8665, Media Watermarking, Security, and Forensics 2013, 86650O (22 March 2013); doi: 10.1117/12.2001563
Show Author Affiliations
Jan Kodovský, Binghamton Univ., SUNY (United States)
Jessica Fridrich, Binghamton Univ., SUNY (United States)

Published in SPIE Proceedings Vol. 8665:
Media Watermarking, Security, and Forensics 2013
Adnan M. Alattar; Nasir D. Memon; Chad D. Heitzenrater, Editor(s)

© SPIE. Terms of Use
Back to Top