FSDCS-Net: A Neural Network for Extracting Residuals of Steganographic Images in the DCT Domain

This repository contains the source code for a research project that is currently being written as a paper. The code and related materials in this repository are intended solely for use in support of the research and educational activities associated with the project, and may not be used or copied for any other purposes without express permission from the authors.

Please note that the paper has not yet been accepted for publication, and the contents of this repository should therefore be considered provisional and subject to change. We welcome feedback and suggestions, but please do not distribute any part of this repository without prior consent from the authors.

Thank you for your understanding and cooperation.

Authors

• Pan Zeng

Installation

We support python3. To install the dependencies run:

pip install -r requirements.txt

Input & Output

Input : A set of stego JPEG images

Output : A location map that indicates the embedding positions in the input images, represented by a binary matrix with the same size as the input images

Dataset

Bossbase. This dataset can be download here.

Additional notes

Citation: