Real-time Student Class Attendance System with AI Face Recognition

Project Description

This project aims to develop a comprehensive end-to-end real-time student class attendance system using AI face recognition. The system operates as follows: After the class concludes, students stand in front of a camera positioned within the classroom. They position their faces within a designated red square frame within the camera's view. When the AI detects a face, it performs facial recognition and crops the image for further processing. The cropped image is then sent to the YOLOv5 model.

The project is divided into two main modules:

1. capture_image.py

   • This module captures images of students for face recognition.
   • Images are saved in the current folder with filenames corresponding to the students' names.

2. face_recognition_code.py

   • This module uses the face_recognition library, which boasts an accuracy of 99.38%, compared to the approximate 95% accuracy of the YOLOv5 model.
   • After recognizing the face, it updates the attendance of the student in an Excel file maintained by this module.

Getting Started

To use this attendance system, follow these steps:

1. Prerequisites:

   • Ensure you have Python installed on your system (Python 3.x recommended).
   • Install the required libraries mentioned in the requirements.txt file using pip install -r requirements.txt.

2. Capture Student Images:

   • Run capture_image.py to capture student images. Enter the student's name when prompted. Images will be saved in the current folder.
   • Hit 'r' on the keyboard to retake the image.
   • Hit 'q' on the keyboard to quit the window and save the image to the current directory.

3. Recognition and Attendance:

   • Run face_recognition_code.py to perform facial recognition and update attendance in the Excel file.
   • The facial recognition process will use the highly accurate face_recognition library.
   • Make sure the camera is set up to capture students' faces as described in the project description.
   • Hit 'q' on keyboard to quit the window.

Project Structure

• capture_image.py: Captures student images for face recognition.
• face_recognition_code.py: Performs face recognition and updates attendance.
• requirements.txt: Lists the required Python libraries for this project.
• attendance.xls: Excel file to maintain attendance records.

Dependencies

This project relies on the following Python libraries:

• face_recognition (requires visual studio c++ compiler, cmake and dlib to be pre-installed)
• OpenCV
• numpy
• xlrd

Please refer to requirements.txt for the exact versions.

License

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