NatnaeAssefa/ethiopian_counterfiet_currency_detection

This final year project focuses on detecting counterfeit currency using deep learning–based image classification. A convolutional neural network model based on VGG16 is trained to classify images of currency notes into their respective denominations and authenticity classes. The trained model is integrated into a Python Tkinter desktop application. ### 🎯 Project Title **Counterfeit Currency Detection Using VGG16 and Image Classification** --- ### 📖 Overview Counterfeit currency poses a significant economic threat. This project presents an automated system that uses **deep learning image classification** techniques to identify counterfeit currency notes. By leveraging **VGG16**, a pre-trained convolutional neural network, the system achieves accurate classification with limited training data. The project consists of: * A **Jupyter Notebook (`.ipynb`)** for model training and evaluation * A **Tkinter-based GUI application (`app.py`)** for real-time testing using trained model weights --- ### 🧠 Technologies Used * Python 3 * TensorFlow / Keras * OpenCV * NumPy * Pillow (PIL) * Tkinter (GUI) * Pretrained **VGG16** model --- ### 🏗️ System Architecture 1. **Dataset Preparation** * Images of currency notes are collected and labeled by denomination/authenticity. 2. **Model Training** * VGG16 is used as a feature extractor. * Custom classification layers are added and trained. 3. **Model Export** * Trained model is saved as a `.h5` file. 4. **GUI Application** * Users load the model weights. * Upload a currency image. * The system preprocesses the image and predicts the class. --- ### 🖥️ GUI Application Features * Browse and load trained `.h5` model weights * Upload and preview currency images * Automatic image preprocessing * Real-time prediction display * Simple and user-friendly interface --- ### 📂 Project Structure ``` ├── app.py # Tkinter GUI application ├── final_project.ipynb # Model training and evaluation notebook ├── model_weights.h5 # Trained model (generated after training) ├── dataset/ # Currency image dataset └── README.md ``` --- ### ▶️ How to Run the Project #### 1️⃣ Install Dependencies ```bash pip install tensorflow opencv-python pillow numpy ``` #### 2️⃣ Train the Model * Open `final_project.ipynb` * Run all cells to train the model * …