DeepShield is a deep learning based web application that detects whether an image or video is AI-generated or real.

It uses a fine-tuned MobileNetV2 model trained on real vs AI-generated datasets and provides security in :

• ✅ AI / Real classification
• 📊 Confidence percentage meter
• 🎨 Red (AI) / Green (Real) result indication
• 📅 Date & time of analysis
• 🎥 Video frame analysis support
• 🚀 Modern React frontend + Flask backend

1. User uploads image or video
2. Backend preprocesses media and recogniase image
3. Model predicts probability
4. Smart threshold logic determines label
5. Result displayed with confidence meter

• React js
• Axios
• CSS3
• Responsive UI

• Flask
• TensorFlow / Keras
• OpenCV
• NumPy

• MobileNetV2 (Pretrained on ImageNet)
• Transfer Learning + Fine Tuning
• Data Augmentation

======= 🛡️ DeepShield – AI Image & Video Detection System

DeepShield is a deep learning-powered web application that detects whether an image or video is AI-generated or real . Built with a modern full-stack architecture , DeepShield combines computer vision, deep learning, and web technologies to provide a fast and intuitive detection system .

🚀 Live Demo 🌐 Frontend (Vercel): https://deep-shield-ai-image-detector.vercel.app ⚙️ Backend (Render): https://deepshield-ai-image-detector-1.onrender.com

✨ Features ✅ AI vs Real classification 📊 Confidence score with probability meter 🎨 Color-based result: 🔴 Red → AI Generated 🟢 Green → Real 📅 Timestamp of analysis

🎥 Video frame-by-frame analysis ⚡ Fast API response with optimized model 📱 Fully responsive UI

🧠 How It Works User uploads an image or video Backend preprocesses the media using OpenCV Frames/images are normalized and resized Model predicts probability using MobileNetV2 Smart threshold logic determines: AI Generated Real Result is returned with confidence score and displayed on UI

🏗️ Tech Stack

🎨 Frontend React.js Axios CSS3 Responsive Design

⚙️ Backend Flask TensorFlow / Keras OpenCV NumPy

🤖 Model MobileNetV2 (Pretrained on ImageNet) Transfer Learning Fine-tuning Data Augmentation

📂 Project Structure

DeepShield/ │ ├── backend/ │ ├── app.py │ ├── train_model.py │ ├── models/ │ ├── uploads/ │ └── requirements.txt │ ├── frontend/ │ ├── src/ │ ├── public/ │ └── package.json │ └── README.md ⚙️ Backend Setup cd backend pip install -r requirements.txt python app.py

🔗 Backend runs at:

http://localhost:5000 🎨 Frontend Setup cd frontend npm install npm start

🔗 Frontend runs at:

http://localhost:3000 🧪 Model Training

To retrain the model:

python train_model.py 📁 Dataset Structure dataset/ │ ├── train/ │ ├── real/ │ └── fake/ │ ├── validation/ │ ├── real/ │ └── fake/ 📦 Backend Requirements

Create backend/requirements.txt:

Flask==3.0.2 flask-cors==4.0.0 tensorflow==2.15.0 numpy==1.26.4 opencv-python==4.9.0.80 Pillow==10.2.0

✅ Compatible with Python 3.10 / 3.11

📦 Frontend Dependencies

Inside frontend/package.json:

"dependencies": { "axios": "^1.6.7", "react": "^18.2.0", "react-dom": "^18.2.0", "react-router-dom": "^6.21.2", "react-scripts": "5.0.1" }

Then run:

npm install ☁️ Deployment 🔹 Frontend Platform: Vercel Auto-deploy from GitHub 🔹 Backend Platform: Render Flask API deployed with environment configuration 📈 Future Improvements 🔍 Larger dataset (10K+ images) 🤖 Ensemble models for higher accuracy 🔥 Explainable AI (Grad-CAM heatmaps) ☁️ Full cloud deployment (AWS / Azure) 🔐 User authentication system 📊 Analytics dashboard 👩‍💻 Author

Tanavi Shinde 🎓 BSc IT Student 💡 AI | DevOps | Full Stack Enthusiast

Building intelligent systems with modern architecture 🚀

⭐ Support

If you like this project: ⭐ Star the repository 🍴 Fork it 🛠️ Contribute

📜 License

This project is open-source and available under the MIT License.