Downloads

Steganography

Steganographic Algorithms

Matlab, MEX a C++ implementations of selected steganographic algorithms for spatial domain, JPEG and side-informed JPEG.

Syndrome Trellis Codes Toolbox

C++ and Matlab implementation of the Syndrome-Trellis Codes with the Multi-layered construction

Steganography Design

C++ implementation of the embedding algorithms from the paper "Design of Adaptive Steganographic Schemes for Digital Images"

Gibbs Construction in Steganography

C++ implementation of the embedding algorithms following the Gibbs construction

Simulator of nsF5

Matlab implementation of the optimally coded nsF5 algorithm

Perturbed Quantization

Matlab implementation of the perturbed quantization steganography

Steganalysis

Breaking ALASKA: Color Separation for Steganalysis in JPEG Domain

Tensorflow/python implementation and saved models

Feature Extractors

Matlab implementations of selected feature extractors for both JPEG and spatial domain steganalysis

Ensemble classifier

Matlab implementation of the ensemble classifier for steganalysis in high dimensions and large training sets

Low-complexiy Linear Classifier

Matlab implementation of the low-complexity linear classifier for steganalysis in high dimensions and large training sets

LRT Linear Classifier

Matlab implementation of the LRT linear classifier for steganalysis in high dimensions and large training sets

Histogram Layer

C++/CUDA implementation of the histogram layer using Caffe framework for steganalysis

Explicit Feature Maps

Matlab implementation of non linear transformation of features

Extractor of 274/548 Merged Features

Matlab and C++ implementations of the JPEG image extractor of 274 merged extended DCT and Markov features (and its 548-dimensional Cartesian calibrated version)

Structural LSB Detectors

Matlab implementation of four quantitative LSB detectors: weighted-stego (WS), sample pairs (SP), triples analysis (T), and Asymptotically Uniformly Most Powerful (AUMP) detector

Quantitative Steganalysis Using Rich Models

Matlab implementation of the quantitative-steganalysis framework utilizing high-dimensional feature spaces (rich models).

Advancing the JPEG Compatibility Attack: Theory, Performance, Robustness, and Practice

Supporting document and stego images relevant to the paper.

Digital Forensics

Matlab implementation of digital camera fingerprint extraction, removal of non-unique artifacts from fingerprints, extraction of noise residual from images, and calculation of Peak-correlation-to-correlation-ratio (PCE) detection statistic.