This projects implements deep steganography techniques and integrates them with existing steganography methods. The project is designed to hide messages within images and extract them to ensure visual similarity and security between original and embedded images. The project uses Convulational Neural Networks and uses three existing image steganography methods which are Least Significant Bit (LSB), Discrete Cosine transform (DCT) and Spatial Universal Wavelet Relative distortion (S-UNIWARD). The projects compares between the original methods and the methods integrated with deep learning to answer the research question while also meeting the objectives stated in the Dissertation Document.
Rutendo Beverly Musara
Student ID: 9733633
The main enviroment used for this project was Jupyter Notebook
I would like to acknowledge Coventry University as this Dissertation is submitted on the behalf of this Univeristy
Dr Norlaily Yaacob - Project Supervisor who helped me with this project
Liang Wang - Developed this dataset used for this project
Figshare - Obtained dataset from this database platform
The dataset used for this Disseration was licensed by CC BY 4.0 DEEDS
Copyright (c) Project Jupyter Contributors. Distributed under the terms of the 3-Clause BSD License.
This project is licensed under the MIT license
LICENCES FOR THE STEGANOGRAPHY TOOLS USED IN THE PROJECT Aperisolve And