Currency Authenticity Detector

A machine learning-based web application that detects whether a currency note is genuine or counterfeit using image classification with a Convolutional Neural Network (CNN). Built with TensorFlow, Keras, OpenCV, and Streamlit.

Features

• Upload image of a currency note
• Predicts whether the note is genuine or fake
• Confidence score displayed with styled results
• CNN model trained from scratch using Keras
• Responsive and interactive frontend built using Streamlit

How It Works

1. The CNN model is trained using hundreds of real and fake currency images.
2. It learns key visual features that differentiate authentic notes from counterfeit ones.
3. Once trained, the model is saved in .keras format.
4. The Streamlit frontend allows users to upload an image, which is processed and passed to the model for prediction.
5. The app shows whether the note is real or fake, with a confidence score and supporting details.

Sample Prediction Flow

1. Upload an image (.jpg, .jpeg, or .png) of a currency note.
2. App resizes the image to 224x224, normalizes it, and feeds it to the model.
3. The model returns a confidence score.
4. A styled box displays whether the note is real or counterfeit.

Installation

Clone the repository

git clone https://github.com/aadyatalreja/counterfeit-currency-checker.git
cd counterfeit-currency-checker

Install required packages

pip install -r requirements.txt

Launch the web app

streamlit run app.py

Dataset

The dataset should be organized as follows:

/IndianCurrencyDataset/
├── train/
│   ├── real/
│   └── fake/
├── validation/
│   ├── real/
│   └── fake/
├── test/
    ├── real/
    └── fake/

Disclaimer

This tool is for educational purposes only. It is not intended for commercial or financial decision-making. Always verify results with official sources or experts.