Development Of Face Recognition Application For Quiz Login System Using Convolutional Neural Network

Huanfareyzo, Waezqorney (2025) Development Of Face Recognition Application For Quiz Login System Using Convolutional Neural Network. Undergraduate thesis, Politeknik Negeri Jember.

	Text (Bab 1 Pendahuluan) Pendahuluan_Skripsi_WaezqorneyHuanfareyzo.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (171kB)
	Text (Ringkasan / Abstract) Abstract_Skripsi_WaezqorneyHuanfareyzo.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (252kB)
	Text (Daftar Pustaka) Daftar-pustaka_Skripsi_WaezqorneyHuanfareyzo.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (277kB)
	Text (Laporan Lengkap) Final_Format_Polije_Skripsi_WaezqorneyHuanfareyzo.pdf - Submitted Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

This study developed a mobile application with supporting web platform for managing and delivering educational content, to enhance academic integrity a face recognition using MobileNetV2 and real-time blink detection were implemented with Flutter and Laravel. The authentication system has two factor pipelines, which are convolutional login and face scan to address identity impersonation. Utilizing of 390 face images from six individuals, the MobileNetV2 model achieved a validation accuracy of 98.72% after 50 epochs, with a minor 8% misclassification. Comparison to others models have been conducted revealed that the MobileNetV2 surpassed others model in terms of performance and computational cost, while blink detection effectively mitigated spoofing risks. A questionnaire revealed 46.7% of respondents considered face recognition as very important feature for preventing identity impersonation, with 70% preferring mobile platforms.

Actions (login required)

View Item