Publication:
Face recognition attendance management system (frams) using deep learning

Loading...
Thumbnail Image
Date
2022-07-01
Authors
Saw, Yang Yi
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Abstract
Face Recognition Attendance Management System (FRAMS) is a system that uses facial recognition to track attendance. Face recognition is a way of matching the captured picture identified by the camera to a database in terms of facial attributes. Because of the Covid-19 outbreak, all face-to-face lectures in schools and universities have been replaced with online-based classes, and standard methods of collecting attendance are not suited for use in online-based classes. These circumstances have created significant hurdles for the whole society and organisation. To solve the challenges, the thesis proposes CNN-based FRAMS. The face detection development stage and the face recognition development stage are the two stages of advancement in this project's development. During the face detection development stage, a few approaches are chosen and their performance in terms of face detection accuracy and detection time is examined. MTCNN is the best method to implement in FRAMS due to its good performance by giving 100% face detection accuracy, although its detection time is slower. The pre-trained VGG-16 model is used to carry out transfer learning throughout the face recognition development stage and gives an accuracy of 99%.In conclusion, FRAMS has been developed and able to recognise the face feature of students with extremely few misclassification mistakes.
Description
Keywords
Citation