Header logo is

The Influence of the Image Basis on Modeling and Steganalysis Performance

2010

Conference Paper

ei


We compare two image bases with respect to their capabilities for image modeling and steganalysis. The first basis consists of wavelets, the second is a Laplacian pyramid. Both bases are used to decompose the image into subbands where the local dependency structure is modeled with a linear Bayesian estimator. Similar to existing approaches, the image model is used to predict coefficient values from their neighborhoods, and the final classification step uses statistical descriptors of the residual. Our findings are counter-intuitive on first sight: Although Laplacian pyramids have better image modeling capabilities than wavelets, steganalysis based on wavelets is much more successful. We present a number of experiments that suggest possible explanations for this result.

Author(s): Schwamberger, V. and Le, PHD. and Schölkopf, B. and Franz, MO.
Book Title: Information Hiding
Journal: Information Hiding:12th International Conference (IH 2010)
Pages: 133-144
Year: 2010
Month: June
Day: 0
Editors: R B{\"o}hme and PWL Fong and R Safavi-Naini
Publisher: Springer

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

DOI: 10.1007/978-3-642-16435-4_11
Event Name: 12th international Workshop (IH 2010)
Event Place: Calgary, AB, Canada

Address: Berlin, Germany
Digital: 0
ISBN: 978-3-642-16435-4
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{6558,
  title = {The Influence of the Image Basis on Modeling and Steganalysis Performance},
  author = {Schwamberger, V. and Le, PHD. and Sch{\"o}lkopf, B. and Franz, MO.},
  journal = {Information Hiding:12th International Conference (IH 2010)},
  booktitle = {Information Hiding},
  pages = {133-144},
  editors = {R B{\"o}hme and PWL Fong and R Safavi-Naini},
  publisher = {Springer},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Berlin, Germany},
  month = jun,
  year = {2010},
  doi = {10.1007/978-3-642-16435-4_11},
  month_numeric = {6}
}