Contactless Automatic Attendance Management System
In today’s world where time is money, people are trying hard to make every
second count. Moreover, in current times where the world is suffering from
COVID, it is very important to maintain social distancing, as advised by the
World Health Organization (WHO), but it is impossible to avoid any physical
contact with biometric fingerprints and cards check-in devices. In hospitals,
industries, offices, and educational institutes where thousands of people work,
their attendance marking is a big issue. Generally, attendance is marked by
calling names, cards check-in, and biometric fingerprint. These all methods have
drawbacks like time consumption, misuse of check-in cards, and physical
disabilities. The solution we propose to all these problems is based on Real-Time
Video Processing Contactless Attendance Management System based on
machine learning and MySQL.
We are using a Convolutional Neural Network for facial recognition. The dataset
used for training the CNN model consists of images of our class fellows. A 4
megapixel IP camera is installed in a classroom that records the video of students
attending the class. The best frame is extracted from the video and faces are
detected and extracted from the selected frame using Haar Cascade and OpenCV.
The extracted faces are passed to the CNN model for face recognition. The
attendance is marked of the recognized student’s faces in the database designed
in MySQL integrated with the CNN model.
Keywords: Face Recognition, OpenCV, Haar Cascade,Face Detection,Machine Learning,CNN
Tools: Google Colab, GPU, Python, MySQL,Jupyter Notebook,IP Camera
Department: Department of Electrical Engineering
Project Team Members
Name |
Email |
Ameer Hamza
|
ameer2018@namal.edu.pk |
Amir Ali
|
amir2018@namal.edu.pk |
Project Poster