Face-Recognition-Attendance-System

Overview

This project implements a face recognition attendance system using VGG19, a popular convolutional neural network architecture, to classify faces. The system is designed to recognize three different persons.

Components

1. Image Capturing:

   • The image_capturing.ipynb notebook captures 100 images of each person using the Haar Cascade algorithm for face detection. These images serve as the training data for the face recognition model.

2. Model Training:

   • The recognition.ipynb notebook trains the VGG19 model using the captured images. It preprocesses the data, fine-tunes the VGG19 architecture, and trains the model to classify the faces of the three specified persons.

3. Python Script:

   • A Python script project.ipynb is provided to perform face recognition in real-time. The script loads the pre-trained VGG19 model, initializes the camera, and captures the faces of individuals.
   • If the recognized name doesn't exist in the CSV file, it will add the name along with the current timestamp into the CSV file.

Files

• image_capturing.ipynb: The notebook for capturing images of individuals using the Haar Cascade algorithm.
• haarcascade_frontalface_default.xml: This XML file contains the pre-trained Haar Cascade classifier for frontal face detection.
• class_names.txt: This file contains the names of the three persons the model is trained to recognize.
• recognition.ipynb: The notebook for training the VGG19 model with the captured images.
• model.h5: This file contains the pre-trained VGG19 model.
• project.ipynb: The Python script for real-time face recognition and attendance tracking.

Usage

1. Ensure you have the necessary dependencies installed. You can install them using pip install -r requirements.txt.
2. Run the image_capturing.ipynb notebook to capture images of each person.
3. Run the recognition.ipynb notebook to train the VGG19 model with the captured images.
4. Finally, run the project.ipynb script: python project.ipynb.
5. The script will open your camera feed. It will recognize faces in real time and update the attendance CSV file accordingly.

Dataset Structure

Datasets/
│
├── Train/
│   ├── venkat/
│   │   ├── image1.jpg
│   │   ├── image2.jpg
│   │   ├── ...
│   │   └── image80.jpg
│   ├── Aadarsh/
│   │   ├── image1.jpg
│   │   ├── image2.jpg
│   │   ├── ...
│   │   └── image80.jpg
│   └── santhosh/
│       ├── image1.jpg
│       ├── image2.jpg
│       ├── ...
│       └── image80.jpg
│
└── Test/
    ├── venkat/
    │   ├── image1.jpg
    │   ├── image2.jpg
    │   ├── ...
    │   └── image20.jpg
    ├── Aadarsh/
    │   ├── image1.jpg
    │   ├── image2.jpg
    │   ├── ...
    │   └── image20.jpg
    └── santhosh/
        ├── image1.jpg
        ├── image2.jpg
        ├── ...
        └── image20.jpg

Results

Accuracy

Loss

Dependencies

• Python 3.x
• OpenCV
• Keras
• TensorFlow
• Numpy
• MatPlotLib

Contributors

• Vedurupaka Venkata Sai
• Ponnuru Aadarsh
• Gayathri Vankadoth

Contributing

We welcome contributions to improve this project! If you'd like to contribute:

• Fork the repository
• Create a new branch (git checkout -b feature/improvement)
• Make your changes and commit them (git commit -am 'Add new feature')
• Push to the branch (git push origin feature/improvement)
• Create a new Pull Request