Abstract:
Face recognition is very important technologies today, especially after extensive
performance upgrades over the past decade, and these systems are now popular in areas
such as security and commerce. In the traditional method, it is difficult to manage a large
number of students in the classroom, since it takes time and has a high risk of error when
entering data into the system, so it is not recommended. But the real difficulty lies in
implementing an accurate, real-time attendance system. Real-time facial recognition is a
convenient way to deal with the large number of students who attend every day. In this
paper, a real-time presence detection system was created, and using the OpenCV library,
the video is read as frames. This was developed this system based on the dlib model using
deep learning techniques, which has an accuracy of 99.37, on the Labeled Faces in the Wild
benchmark dataset, combined with the library simple Face Recognition library in python.
For face detection, we used Histogram of Oriented Gradients (HOG). And for face
recognition, we apply Euclidean distance calculation to identify the input face, our system
is capable of recognizing multiple faces, and it will be a successful way to manage student
attendance.
