Development of A Multi-Client Student Attendance Monitoring System

Citation

Jia, Yi Phang and Low, Kwee Yee and Wong, Hwee Ling (2019) Development of A Multi-Client Student Attendance Monitoring System. International Journal of Recent Technology and Engineering, 8 (3S). pp. 6-11. ISSN 2277-3878

[img] Text
49.pdf - Published Version
Restricted to Repository staff only

Download (524kB)

Abstract

A multi-client student attendance student monitoring system was developed. The attendance system consists of the client and the server. The core functions of the client device are verifying student’s identity for attendance recording and monitoring their presence in class. Haar-feature based cascade classifier for object detection and the Scale Invariant Feature Transform Technique (SIFT) technique were implemented for the face authentication process. This paper highlights a full-fledge system architecture with face-based identification implemented on the Raspberry Pi 2 board as the client alongside with RFID authentication for initial identification. The system also has webpage integration for system management. The accuracy achieved was 84% for face verification and 75% for face recognition. The experimental result showed that the recognition rate was affected by inconsistency of wearing glasses, distance between the face and the webcam, lighting condition and the environmental background. A database was setup to store attendance and student information. It is supported with a web application to view, update and analyze the attendance data.

Item Type: Article
Uncontrolled Keywords: Human face recognition (Computer science), Attendance system, embedded system, face recognition, web application.
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Divisions: Faculty of Engineering (FOE)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 06 Sep 2021 14:04
Last Modified: 06 Sep 2021 14:04
URII: http://shdl.mmu.edu.my/id/eprint/8743

Downloads

Downloads per month over past year

View ItemEdit (login required)