RANCANGAN BANGUN SISTEM ABSENSI WAJAH SECURE BIOMETRIC MENGUNAKAN HYBRID ADAFACE DAN MEDIAPIPE SEBAGAI UPAYA MENCEGAH ANTI - SPOOFING

CHAIRULSYAH, DIOFAVIAN RAFIF (2026) RANCANGAN BANGUN SISTEM ABSENSI WAJAH SECURE BIOMETRIC MENGUNAKAN HYBRID ADAFACE DAN MEDIAPIPE SEBAGAI UPAYA MENCEGAH ANTI - SPOOFING. S1 thesis, Universitas Mercu Buana Jakarta.

[img]
Preview
 Text (HAL COVER)
Cover.pdf
Download (648kB) | Preview
[img] Text (BAB I)
Bab 1.pdf
Restricted to Registered users only
Download (42kB)
[img] Text (BAB II)
Bab 2.pdf
Restricted to Registered users only
Download (4MB)
[img] Text (BAB III)
Bab 3.pdf
Restricted to Registered users only
Download (499kB)
[img] Text (BAB IV)
Bab 4.pdf
Restricted to Registered users only
Download (5MB)
[img] Text (BAB V)
Bab 5.pdf
Restricted to Registered users only
Download (118kB)
[img] Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf
Restricted to Registered users only
Download (99kB)
[img] Text (LAMPIRAN)
Lampiran.pdf
Restricted to Registered users only
Download (547kB)

Abstract

This research develops a web-based Secure Biometric facial attendance system to address the weaknesses of conventional systems that are vulnerable to attendance manipulation (spoofing) and accuracy degradation in low-resolution imagery. The system proposes a hybrid approach combining the AdaFace model for identity recognition based on Quality-Adaptive Margin and MediaPipe Face Mesh for realtime facial landmark detection. Key security features include Active Liveness Detection with randomized head direction challenges and smile detection, Baseline Calibration mechanism for relative movement accuracy, MiniFASNetV2 for detecting fake faces from photos or videos, and electronic device detection using YOLOv11n to prevent screen-based attacks. Testing was conducted through an experimental study involving 40 subjects with demographic variations and 180 Presentation Attack scenarios. Test results demonstrate that the system achieves 100% recognition accuracy for legitimate users and 86% Success Rate in intensive reliability testing. The system also exhibits superior resistance to spoofing attacks with 100% Prevention Rate for printed photos, digital photos, and video replay attacks, as well as 73.3% for video call streaming attacks, resulting in an overall security rate of 94.7%. This research concludes that the integration of hybrid methods effectively creates a secure, accurate, and reliable attendance system for real-world implementation. Keywords: Facial Attendance, AdaFace, MediaPipe, MiniFASNet, YOLOv11n, Liveness Detection, Anti-Spoofing, Presentation Attack, MediaPipe, Deep Learning. Penelitian ini mengembangkan sistem absensi wajah Secure Biometric berbasis web untuk mengatasi kelemahan sistem konvensional yang rentan terhadap manipulasi kehadiran (spoofing) dan penurunan akurasi pada citra resolusi rendah. Sistem mengusulkan pendekatan hybrid yang menggabungkan model AdaFace untuk pengenalan identitas berbasis Quality-Adaptive Margin dan MediaPipe Face Mesh untuk deteksi landmark wajah secara real-time. Fitur keamanan utama meliputi Active Liveness Detection dengan tantangan arah kepala acak dan deteksi senyum, mekanisme Baseline Calibration untuk akurasi gerakan relatif, MiniFASNetV2 untuk deteksi wajah palsu dari foto maupun video, serta deteksi perangkat elektronik menggunakan YOLOv11n untuk mencegah serangan berbasis layar. Pengujian dilakukan melalui studi eksperimental yang melibatkan 40 subjek dengan variasi demografi serta 180 skenario serangan Presentation Attack. Hasil pengujian menunjukkan bahwa sistem mencapai tingkat akurasi pengenalan 100% pada pengguna sah dan Success Rate 86% dalam uji reliabilitas intensif. Sistem juga menunjukkan ketahanan superior terhadap serangan spoofing dengan Prevention Rate 100% pada serangan foto cetak, foto digital, dan video replay, serta 73,3% pada serangan video call streaming, menghasilkan tingkat keamanan keseluruhan sebesar 94,7%. Penelitian ini menyimpulkan bahwa integrasi metode hybrid efektif menciptakan sistem absensi yang aman, akurat, dan andal untuk implementasi dunia nyata. Kata Kunci: Absensi Wajah, AdaFace, MediaPipe, MiniFASNet, YOLOv11n, Liveness Detection, Anti-Spoofing, Presentation Attack, MediaPipe, Deep Learning

Item Type: Thesis (S1)
NIM/NIDN Creators: 41522010091
Uncontrolled Keywords: Absensi Wajah, AdaFace, MediaPipe, MiniFASNet, YOLOv11n, Liveness Detection, Anti-Spoofing, Presentation Attack, MediaPipe, Deep Learning
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 006 Special Computer Methods/Metode Komputer Tertentu > 006.3 Artificial Intelligence/Kecerdasan Buatan > 006.31 Machine Learning/Pembelajaran Mesin
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: khalimah
Date Deposited: 31 Jan 2026 06:29
Last Modified: 31 Jan 2026 06:29
URI: http://repository.mercubuana.ac.id/id/eprint/100799

Actions (login required)

View Item View Item
UMB Repository 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.