A FACIAL RECOGNITION SYSTEM USING A COMBINATION OF YOLOV8, ARCFACE, AND SVM FOR NON-CONTACT EMPLOYEE ATTENDANCE (Case Study: GKST Sinar Kasih Hospital, Tentena)

TERAMPE, GLANES CINDY (2026) A FACIAL RECOGNITION SYSTEM USING A COMBINATION OF YOLOV8, ARCFACE, AND SVM FOR NON-CONTACT EMPLOYEE ATTENDANCE (Case Study: GKST Sinar Kasih Hospital, Tentena). Tugas Akhir thesis, University of Technology Yogyakarta.

[img] Text
5220411219_Glanes Cindy Terampe_ABSTRAK.pdf
Download (157kB)

Abstract

ABSTRACT This research develops an automatic attendance system based on facial recognition to address the limitations of manual systems, which are prone to manipulation and time-consuming. The primary focus is to optimise the system for surveillance cameras installed at 3 meters, integrating YOLOv8m as the face detector, ArcFace for 512-dimensional feature extraction, and a polynomial kernel SVM for identification. The dataset comprises five individuals and is augmented 20-fold per image, then split into training (70%) and test (30%) sets. The preprocessing stage includes CLAHE, denoising, and sharpening to enhance image quality. Evaluation results demonstrate competitive performance, with an accuracy of 93.7%, precision of 0.938, recall of 0.937, and an F1 score of 0.935. PCA and confusion matrix analyses reveal that the primary misclassification occurs between two classes with high feature similarity: E005_employee5 and E002_employee2. Regarding efficiency, the system achieves 7.2 FPS on the test device. These results indicate that the system is both accurate and practical for real-world applications, although real-time performance depends heavily on the hardware used. Keywords: Deep Learning; Facial Recognition; YOLOv8; ArcFace; Support Vector Machine.

Item Type: Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects: T Technology > T Technology (General) > T201 Patents. Trademarks
Divisions: Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User: Kaprodi S1 Informatika UTY
Date Deposited: 07 May 2026 02:56
Last Modified: 07 May 2026 02:56
URI: http://eprints.uty.ac.id/id/eprint/19813

Actions (login required)

View Item View Item
UTY Open Access is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.