PENGENALAN WAJAH UNTUK AUTENTIKASI LOGIN MAHASISWA MENGGUNAKAN YOLOV8 DAN INCEPTIONRESNETV1 PRETRAINED PADA DATASET VGGFACE2 BERBASIS WEBSITE

Gunawan, Ellexia Leonie (2025) PENGENALAN WAJAH UNTUK AUTENTIKASI LOGIN MAHASISWA MENGGUNAKAN YOLOV8 DAN INCEPTIONRESNETV1 PRETRAINED PADA DATASET VGGFACE2 BERBASIS WEBSITE. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) Skrpsi Cover.pdf Download (1MB)
	Text (Bab I) Bab 1.pdf Download (93kB)
	Text (Bab II) Bab 2.pdf Restricted to Repository staff only until 27 May 2028. Download (659kB)
	Text (Bab III) Bab 3.pdf Restricted to Repository staff only until 27 May 2028. Download (551kB)
	Text (Bab IV) Bab 4.pdf Restricted to Repository staff only until 27 May 2028. Download (791kB)
	Text (Bab V) Bab 5.pdf Download (16kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf Download (90kB)
	Text (Lampiran) Lampiran.pdf Restricted to Repository staff only until 27 May 2028. Download (83kB)

Abstract

Password-based authentication remains a major vulnerability in protecting students’ personal data, particularly in web-based login systems. Weak or easily compromised passwords allow unauthorized access to sensitive information. Addressing this issue, this study aims to develop a more secure and efficient student login authentication system based on facial recognition. The proposed system integrates YOLOv8 for face detection and InceptionResNetV1 for feature extraction and identity verification. A local web-based interface was developed using Gradio to support face authentication in a simple and interactive manner. Facial data were collected from 67 undergraduate students of the Data Science program at UPN “Veteran” Jawa Timur under proper lighting conditions. The research methodology included image preprocessing, training of the face detection model using YOLOv8, and facial feature embedding using InceptionResNetV1 with two different pretrained models: VGGFace2 and CASIA-WebFace. This research is urgent as it responds to the growing need for secure authentication methods tailored for academic information systems. The innovation lies in the end-to-end integration of a pretrained InceptionResNetV1 model with fast YOLOv8-based face detection within a web environment, distinguishing it from prior studies that often focus on isolated components or lack web implementation. Evaluation results show that the model pretrained with VGGFace2 outperformed CASIA-WebFace, achieving the best results at a 0.6 threshold: 98.75% accuracy, 98.53% precision, 100% recall, and 99.26% F1-score. These outcomes demonstrate the effectiveness and feasibility of the developed system for enhancing student login security through facial recognition.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Diyasa, I Gede Susrama Mas	NIDN0019067008	igsusrama.if@upnjatim.ac.id
Thesis advisor	Saputra, Wahyu Syaifullah Jauharis	NIDN0725088601	wahyu.s.j.saputra.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Ellexia Leonie Gunawan

Date Deposited:

27 May 2025 06:47

Last Modified:

27 May 2025 06:47

URI:

https://repository.upnjatim.ac.id/id/eprint/36731