OpenCV Project: Student Attendance and Facial Expression Detection - My Project LKTIM 2020

Overview

• This project combines computer vision and deep learning techniques to facilitate student attendance and facial expression detection. Using OpenCV for real-time face detection and TensorFlow/Keras for emotion recognition, the system provides two main functionalities: training a model for facial expression recognition and performing real-time emotion detection via a webcam feed. It is specifically designed to track student attendance based on facial expressions and can operate either by training a recognition model or by displaying live emotion detection results.

• This is my project (LKTIM 2020 --> Lomba Karya Tulis Ilmiah Mahasiswa 2020).

• National Student Scientific Writing Competition.

Summary:

• This project is designed for student attendance and facial expression detection using OpenCV and TensorFlow. The system can either train a facial expression recognition model or display real-time emotion detection using a webcam feed.

Directory Structure / Explanation :

Student_Attendance_and_Facial_Expression_Detection_Project/
├── data/ # Directory for training and testing data.
│ ├── test/ # Testing data.
│ └── train/ # Training data.
├── dataSet/ # Directory for dataset files.
├── recognizer/ # Directory for face recognition training_data.yml
├── trainer/ # Directory for face recognition training_data.yml
├── dataset.py # Script for managing datasets.
├── datatrainner.py # Script for training the model.
├── emotions.py # Script for real-time emotion detection and attendance.
├── face.csv # Tracks student attendance using facial recognition data and numbered dataset codes.
├── haarcascade_frontalface_default.xml # Haar Cascade XML file for face detection.
├── How-to-Run.txt # rovides commands for real-time emotion detection or model training.
├── model.h5 # file contains the saved weights of the trained facial expression recognition model.
├── output[i].avi # Contains a video of the live webcam feed with face and emotion annotations.
├── README.md # Contains purpose, usage instructions, and setup information.
├── requirements.txt # List of required Python packages.
└── time.py # Script for time-related functionalities.

Dependencies

The project requires the following Python packages. You can install them using pip:

Features

• Facial Expression Detection: Uses a convolutional neural network (CNN) to detect and classify facial expressions into seven categories.
• Student Attendance: Captures and records student attendance based on facial recognition.
• Real-time Processing: Displays live video feed with detected emotions and student names.

---

Requirements

• Python 3.x
• Libraries listed in requirements.txt =
  • numpy
  • argparse
  • opencv-python
  • Pillow
  • tensorflow
  • matplotlib
  • imutils

---

Installation

1. Clone the repository:

   git clone <repository-url>
   cd <repository-directory>

2. Install the required packages:

   Create a requirements.txt file with the following content:

   numpy
   argparse
   opencv-python
   Pillow
   tensorflow
   matplotlib
   imutils
   

   Then, install the packages using:

   pip install -r requirements.txt

---

Usage

Training the Model

To train the facial expression recognition model, use the following command:

python emotions.py --mode train

This will:

• Train the model using images from the data/train and data/test directories.
• Save the trained model weights to model.h5.
• Plot and save the training history as plot.png.

Display Mode

To start the real-time emotion detection and student attendance system, use:

python emotions.py --mode display

This will:

• Open the webcam and display the live video feed.
• Detect faces and predict emotions from the webcam feed.
• Record attendance in face.csv if the user inputs the correct class when prompted.

---

Files

• emotions.py: Main script for training and displaying.
• haarcascade_frontalface_default.xml: Haar cascade for face detection.
• recognizer/training_data.yml: Pre-trained data for face recognition (required for display mode).
• data/train and data/test: Directories containing training and validation images.

---

Interfaces:

• Detect expressions of students whose data has been registered:
  
  

• Detect expressions of students whose data has not been registered:
  
  

---

Accuracy Testing

• 
  

Notes

• Ensure that the haarcascade_frontalface_default.xml file is in the same directory as emotions.py.
• Modify the script emotions.py to match your specific face recognizer setup and class IDs.
• The dataset.py and datatrainner.py scripts should be used as required for dataset preparation and model training, respectively.

---

License

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

@Copyright 2020 | Team-LKTI-PSMURO-GUNDAR