This is a web-based demonstration app for detecting and classifying Indian currency notes using a Roboflow-trained instance segmentation model. The app enables real-time inference directly from uploaded images or webcam input, labeling each note with its denomination.
🎯 Model URL: Currency Recognition Model - Roboflow 🎯 Live Demo URL: Currency Recognition Model - Git Page
- 🧠 Real-time inference using Roboflow API
- 📷 Upload images or use your webcam
- 🧾 Detects and labels Indian currency notes
- 🔲 Instance segmentation with bounding boxes
- ⚡️ Fully browser-based (no backend server needed)
- Model Type: Instance Segmentation
- Platform: Roboflow
- Classes:
- ₹0, ₹1, ₹2, ₹5, ₹10, ₹20, ₹50, ₹100, ₹200, ₹500
- Use Case: Currency recognition for education, accessibility tools, low-vision user aid, kiosk automation, etc.
🌐 GitHub Pages link (Click-here)
You can test this model directly from your browser using:
- Uploaded images
- Live webcam feed
git clone https://github.com/yourusername/currency-recognition-webapp.git
cd currency-recognition-webappSimply open index.html in your browser:
start index.html # For Windows
open index.html # For macOS
xdg-open index.html # For Linuxcurrency-recognition-webapp/
├── index.html # Main webpage
├── script.js # JS logic (Roboflow API + webcam)
├── style.css # Basic styling
├── sample-image.png # Example image (optional)
└── README.md # This file
Replace the placeholder API key in script.js with your actual Roboflow API key:
const model = await roboflow.load({
model: "currency-recognition-model",
version: 1,
api_key: "YOUR_ROBOFLOW_API_KEY"
});Get your API key from: https://app.roboflow.com
You can host this app for free using GitHub Pages:
- Push the repo to your GitHub account
- Go to Settings → Pages
- Select the branch (e.g.,
main) and root folder - Save and access your live demo via the generated URL
- Roboflow – For providing tools to build and host the model
- Daniel Raj Christeen – Developer of this demo project
This project is licensed under the MIT License – see the LICENSE file for details.
Feel free to open issues, fork, or create pull requests to enhance this project!
Let me know if you'd like me to:
- Generate the
index.html,script.js, andstyle.csstemplates - Help you deploy it directly to GitHub Pages
- Customize the README for academic portfolio or LinkedIn use
I'm happy to help!