Enhancing Security in Deepfake Detection Systems Through Hybrid ECDSA Cryptographic Technique

This video presents our project “Enhancing Security in Deepfake Detection Systems using Hybrid ECDSA (ECC and DSA) Cryptographic Techniques.”

With the rapid rise of deepfake technology, verifying the authenticity of digital media has become a major challenge. Traditional deep learning models can detect fake content, but they do not ensure that the media remains unaltered after verification.

In this project, we propose a hybrid framework that combines:

1. Deep learning-based deepfake detection using ResNet-50

2. Cryptographic verification using ECDSA and SHA-256

🔍 The system is capable of:
Detecting deepfake images and videos

Generating tamper-proof digital signatures

Verifying content integrity and authenticity

Providing a composite trust score

Maintaining secure audit logs for transparency

💡 This approach ensures both:
✔ Accurate deepfake detection

✔ Secure and trustworthy media verification

📊 Applications:

Social media monitoring
Digital forensics
Cybersecurity systems
Law enforcement & evidence validation

🚀 Technologies Used
Deep Learning (ResNet-50)
Python, TensorFlow/Keras
ECDSA (Elliptic Curve Cryptography)
SHA-256 Hashing
Web-based Dashboard Interface

Видео Enhancing Security in Deepfake Detection Systems Through Hybrid ECDSA Cryptographic Technique канала Shriram