Rancang Bangun Sistem Absensi Mahasiswa Otomatis Berbasis Desktop dengan Pendekatan Face Recognition dan Adaptive Attendance monitoring

Imam, Afandy (2026) Rancang Bangun Sistem Absensi Mahasiswa Otomatis Berbasis Desktop dengan Pendekatan Face Recognition dan Adaptive Attendance monitoring. Undergraduate thesis, UPN Veteran JawaTimur.

	Text (Cover) 21081010290-cover.pdf Download (1MB)
	Text (BAB 1) 21081010290-bab1.pdf Download (222kB)
	Text (BAB 2) 21081010290-bab2.pdf Restricted to Repository staff only until 7 January 2029. Download (565kB)
	Text (BAB 3) 21081010290-bab3.pdf Restricted to Repository staff only until 7 January 2029. Download (1MB)
	Text (BAB 4) 21081010290-bab4.pdf Restricted to Repository staff only until 7 January 2029. Download (2MB)
	Text (BAB 5) 21081010290-bab5.pdf Download (208kB)
	Text (Daftar Pustaka) 21081010290-daftarpustaka.pdf Download (216kB)
	Text (Lampiran) 21081010290-lampiran.pdf Restricted to Repository staff only until 7 January 2029. Download (424kB)

Abstract

Manual attendance processes in higher education often face constraints regarding time inefficiency and vulnerability to data manipulation, such as proxy attendance (titip absen). Although Face Recognition technology has begun to be adopted, most existing systems only apply a "once recognition" method, which allows students to leave the class after the initial presence is recorded without validating the full duration of their attendance. This study aims to design and build an automatic student attendance system based on a desktop application controlled by the lecturer, integrating Face Recognition technology and the Adaptive Attendance Monitoring (AAM) method. The system is developed using the Multi-Task Cascaded Convolutional Neural Network (MTCNN) algorithm for face detection and FaceNet for feature embedding extraction, accelerated by CUDA technology on the GPU to ensure real-time performance. The system architecture utilizes a ClientServer model based on REST API, where the client side is constructed using the PySide6 framework and the server side utilizes Flask with MySQL as the database. The AAM method is applied to continuously calculate the accumulated duration of student presence within the camera frame, where a student is declared present only if they meet the minimum duration threshold of 80% of the total lecture session. Test results demonstrate that the system is capable of detecting and recognizing student faces accurately under various lighting conditions and facial positions. The Adaptive Attendance Monitoring feature proved effective in monitoring student participation, automatically resetting the attendance status if a student leaves the camera range beyond the designated tolerance limit, as well as providing transparent and objective attendance recapitulation. This system offers an attendance solution that is more efficient, accurate, and possesses higher integrity compared to conventional methods or standard biometric systems.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Akbar, Fawwaz Ali	NIP.199203172018031002	fawwaz_ali.fik@upnjatim.ac.id
Thesis advisor	Mumpuni, Retno	NIP1987077162025212045	retnomumpuni.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics
Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Imam Afandy

Date Deposited:

07 Jan 2026 06:39

Last Modified:

27 Jan 2026 03:16

URI:

https://repository.upnjatim.ac.id/id/eprint/48587