🥗 Pulse AI

AI-powered food label and ingredient analyzer that helps users understand packaged food products by scanning labels and receiving scientifically grounded insights.

✨ Features

• 📸 Scan and analyze food labels
• 🧠 AI-powered ingredient interpretation
• ⚠️ Detect additives, preservatives, and emulsifiers
• 🥦 Explain nutritional impact of ingredients
• 💬 Context-aware AI chat for follow-up questions
• 📱 Responsive and mobile-friendly UI
• ⚡ Fast analysis with structured JSON responses

---

🚀 Demo

Add your deployed application link here:

Frontend: https://github.com/surajgupt01/NutriScan-AI
Backend:  https://github.com/surajgupt01/PulseBackend

---

🛠️ Tech Stack

Frontend

• Next.js 15
• React
• TypeScript
• Tailwind CSS
• Framer Motion
• Axios

Backend

• FastAPI
• Python
• Pydantic
• Uvicorn

AI

• Google Gemma API (Hugging Face)
• Structured JSON Outputs

---

🏗️ Architecture

User Uploads Image
        │
        ▼
 Next.js Frontend
        │
        ▼
 FastAPI Backend
        │
        ▼
     LLM API
        │
        ▼
 Ingredient Analysis
        │
        ▼
 Interactive Chat UI

---

📂 Project Structure

label-scan-ai/
│
├── frontend/
│   ├── app/
│   ├── components/
│   ├── lib/
│   ├── types/
│   └── public/
│
├── backend/
│   ├── routes/
│   ├── services/
│   ├── models/
│   ├── prompts/
│   └── main.py
│
└── README.md

---

🔧 Installation

Clone the repository

git clone https://github.com/surajgupt01/NutriScan-AI

Frontend Setup

cd frontend

npm install

npm run dev

Frontend runs on:

http://localhost:3000

Backend Setup

cd backend

python -m venv venv

# Linux / Mac
source venv/bin/activate

# Windows
venv\Scripts\activate

pip install -r requirements.txt

uvicorn main:app --reload

Backend runs on:

http://localhost:8000

---

🔑 Environment Variables

Frontend (.env.local)

NEXT_PUBLIC_API_URL=http://localhost:8000/analyze

Backend (.env)

HT_TOEKN=your_api_key

---

💡 How It Works

1. User uploads a food label image.
2. The image is processed by the backend.
3. Gemini analyzes ingredients and additives.
4. Structured JSON is generated.
5. The frontend renders interactive blocks.
6. Users can continue chatting with context.

---

📸 Screenshots

Add screenshots here:

![Home](./screenshots/home.png)
![Scan](./screenshots/scan.png)
![Analysis](./screenshots/analysis.png)

---

🔮 Future Improvements

• Streaming responses
• User authentication
• Scan history
• Barcode scanner integration
• Personalized dietary recommendations
• Ingredient safety scoring
• Multi-language support

---

🤝 Contributing

Contributions are welcome!

git checkout -b feature/my-feature
git commit -m "Add new feature"
git push origin feature/my-feature

Open a Pull Request 🚀

👨‍💻 Author

Suraj Gupta

• GitHub: https://github.com/surajgupt01
• LinkedIn: https://www.linkedin.com/in/suraj-gupta-1894051ba/

If you found this project useful, please give it a ⭐ on GitHub!