Facial Attendance System

Facial Attendance System is a project aimed at automating attendance management using facial recognition. This system detects faces in real-time using a webcam, recognizes the faces of individuals enrolled in the system, and logs their attendance in a CSV file.

Features

• Real-time face detection and recognition using OpenCV and TensorFlow/Keras models (InceptionV3, ResNet50, DenseNet).
• Utilized CNN model, transfer learning with ResNet-50, InceptionV3, and DenseNet-121 to achieve accurate facial recognition.
• Selected DenseNet-121 model for capturing due to its highest accuracy of 96%, increasing accurcy from 84% achieved by the CNN model.
• Streamlit web application for easy interaction with the attendance system.
• Logging attendance records in a CSV file.
• Pickling trained models for easy reusability and deployment.

Requirements

• Python 3.6+
• Libraries: OpenCV, TensorFlow, Keras, Streamlit, NumPy, Pandas, Matplotlib, PIL (Pillow), Scikit-learn.
• Pre-trained models: InceptionV3, ResNet50, DenseNet (or other models suitable for facial recognition).

Installation

1. Clone the repository:

   git clone https://github.com/Prasadayus/Facial-Attendance-System.git
   cd facial-attendance-system

2.Install dependencies:

   pip install -r requirements.txt

Usage

Capturing Images for Training

To capture images for training the facial recognition models:

1.Run the image capture script:

python capture_images.py

2.Follow the instructions to enter the client's name and capture images.

Model Training

To train the facial recognition models:

1.Ensure captured images are stored in the imgs/ directory (not provided in the repository due to privacy reasons).

2.Train the model using TensorFlow/Keras:

python train_model.py

Select the appropriate model architecture and follow the training process.

Logging Attendance

To log attendance using the Streamlit web application:

1.Launch the Flask web application:

python app.py

2.Use the web interface to start the webcam, recognize faces, and log attendance.

3.Attendance records are stored in CSV files located in the logs/ directory.

Contributing

Contributions are welcome! If you have any suggestions, feature requests, or bug reports, please open an issue or submit a pull request.

License

This project is licensed under the MIT License - see the LICENSE file for details.