Pomegranate Disease Detection using CNN

• Introduction

  This project is a web-based application that leverages a Convolutional Neural Network (CNN) to identify common diseases in pomegranates. Users can upload an image of a pomegranate, and the machine learning model will analyze it to detect diseases such as Bacterial Blight, Bitter Rot, and Pomegranate Butterfly.

  The application provides a user-friendly interface for farmers and agricultural professionals to quickly diagnose potential issues with their crops and receive recommendations for treatment.

• Features

  • Utilizes a trained Keras/TensorFlow model to accurately classify pomegranates into four categories:
    1. Bacterial Blight
    2. Bitter Rot
    3. Pomegranate Butterfly
    4. Healthy
  • It also displays suggested pesticides and fungicides based on the prediction.

• Technologies Used

  • Backend: Python, Flask
  • Machine Learning: TensorFlow, Keras, Scikit-learn
  • Image Processing: OpenCV, Pillow
  • Frontend: HTML, CSS, Bootstrap 5, JavaScript
  • Data Handling: NumPy, Matplotlib

• Setup and Installation

  • To get this project running on your local machine, follow these simple steps.

    1. Clone the Repository git clone https://github.com/kidou16/Final-Year-Project.git

       cd Final-Year-Project

    2. Create a Virtual Environment

       • For macOS/Linux

         python3 -m venv venv source venv/bin/activate

       • For Windows

         python -m venv venv venv\Scripts\activate

    3. Install Dependencies (If pip doesn't work try with pip3)

       pip install flask werkzeug opencv-python tensorflow keras numpy matplotlib pillow scikit-learn

    4. Run the Application (By starting Flask development server)

       python3 mySite.py

    5. View in Browser (Open the browser of your choice and navigate to:)

       http://127.0.0.1:5000/

• How to Use

  • Navigate to the "Detect Disease" Page: From the home page, click on the "Detect Disease" link.
  • Upload an Image: Click the "Choose Image to Upload" button and select a clear image of a pomegranate. The image will upload and appear on the page automatically.
  • Analyze the Image: Click the "Test" button to process the image with the CNN model.
  • View Results: The prediction will be displayed in the "Test Report" card, with treatment suggestions in the card below.

• Project Structure

  Here is an overview of the key files and directories in the project

:

Final-Year-Project/ 
├── Dataset/                 # Contains the training and testing image data
├── Results/                 # Model performance charts (Accuracy, Loss, etc.)
├── static/
│   ├── css/                 # Custom CSS stylesheets (style.css)
│   └── images/              # Static images for the UI (backgrounds, etc.)
├── templates/
│   ├── home.html            # The landing page
│   ├── image.html           # The main page for uploading and testing
│   ├── info.html            # About the project page
│   └── layout.html          # The base HTML template with navigation and footer
├── upload/                  # Default directory for user-uploaded images
├── cnn_train.py             # Script to train the CNN model (if included)
├── mySite.py                # The main Flask application file
├── supportFile.py           # Helper script containing the prediction logic
└── Trained_model.h5         # The pre-trained Keras model file

• Contributing

  Contributions are welcome! If you have suggestions for improvements, please feel free to fork the repository and submit a pull request.

• License

  This project is licensed under the MIT License.