Digital Watermarking Tool

10/2022 - 11/2022

Project description

A digital watermarking tool developed for the “Catch the Mark” competition of the Multimedia Data Security graduate course at the University of Trento.

The implementation is based on various research papers and leverages Discrete Wavelet Transform and Singular Value Decomposition.

The repository contains code for embedding, detecting, and attacking watermarked images.

Overall, the embedding strategy was evaluated as interesting and effective.
Azure IaaS (VM with configured Python venv), Google Colab (Jupyter notebooks) and local machines were exploited for the parallelization of the attack phase.

Artifacts

Team and role

Team size: 4 people

• I was responsible for the implementation of the embedding and detection algorithm along with another team member.
  
• I proposed and set up simple but effective cloud services for the attack phase.
  
• Given my previous experience with Python and its main libraries, I took the lead for the supervision of the project and the coordination of the team.
• Since our backgrounds were pretty different, we tried to leverage the most of each other’s skills and knowledge, from telecommunications to software engineering and cybersecurity.
  

Tech stack