Banknote Classification System

This project develops an advanced deep learning solution for the real-time recognition and classification of Bangladeshi banknotes. By leveraging Convolutional Neural Networks (CNNs), the system is capable of classifying eight different currency denominations (2, 5, 10, 20, 50, 100, 500, and 1000 Taka) with high accuracy. The solution provides a RESTful API for production integration and a user-friendly web interface using Streamlit for interactive visualizations. With over 95% accuracy and real-time processing, the system includes a fully-trained model, image preprocessing pipeline, and confidence scoring for each prediction. It is designed for scalability, cross-platform compatibility, and can be easily integrated into various applications requiring currency verification or financial systems.