J
Jan Kodovsky
Researcher at Binghamton University
Publications - 7
Citations - 2776
Jan Kodovsky is an academic researcher from Binghamton University. The author has contributed to research in topics: Steganalysis & Steganography. The author has an hindex of 7, co-authored 7 publications receiving 2198 citations.
Papers
More filters
Journal ArticleDOI
Rich Models for Steganalysis of Digital Images
Jessica Fridrich,Jan Kodovsky +1 more
TL;DR: A novel general strategy for building steganography detectors for digital images by assembling a rich model of the noise component as a union of many diverse submodels formed by joint distributions of neighboring samples from quantized image noise residuals obtained using linear and nonlinear high-pass filters.
Journal ArticleDOI
Ensemble Classifiers for Steganalysis of Digital Media
TL;DR: This paper proposes an alternative and well-known machine learning tool-ensemble classifiers implemented as random forests-and argues that they are ideally suited for steganalysis.
Proceedings ArticleDOI
Multivariate gaussian model for designing additive distortion for steganography
Jessica Fridrich,Jan Kodovsky +1 more
TL;DR: This paper adopts a different strategy in which the cover is modeled as a sequence of independent but not necessarily identically distributed quantized Gaussians and the embedding change probabilities are derived to minimize the total KL divergence within the chosen model for a given embedding operation and payload.
Proceedings ArticleDOI
On dangers of overtraining steganography to incomplete cover model
TL;DR: It is shown that, quite surprisingly, even a high-dimensional cover model does not automatically guarantee immunity to simple attacks and the security can be compromised if the distortion is optimized to an incomplete cover model.
Journal ArticleDOI
Quantitative Structural Steganalysis of Jsteg
Jan Kodovsky,Jessica Fridrich +1 more
TL;DR: Two new classes of quantitative steganalysis methods for the steganographic algorithm Jsteg are proposed, one of which obtains the change-rate estimate using a maximum likelihood estimator equipped with a precover model and the other by minimizing an objective function constructed from a heuristically formed zero message hypothesis.