ALGIFARI, MOHAMAD RODZIYAN (2026) IMPLEMENTASI TEKNOLOGI BIOMETRIK PENGENALAN WAJAH BERBASIS ARCFACE DAN PELACAK LOKASI UNTUK VALIDASI DATA KEHADIRAN PEGAWAI. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (472kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab_1.pdf Restricted to Registered users only Download (153kB) |
|
|
Text (BAB II)
Bab_2.pdf Restricted to Registered users only Download (220kB) |
|
|
Text (BAB III)
Bab_3.pdf Restricted to Registered users only Download (220kB) |
|
|
Text (BAB IV)
Bab_4.pdf Restricted to Registered users only Download (609kB) |
|
|
Text (BAB V)
Bab_5.pdf Restricted to Registered users only Download (135kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar_Pustaka.pdf Restricted to Registered users only Download (28kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (182kB) |
Abstract
Conventional attendance systems still suffer from several security weaknesses. Buddy punching, location fraud, and time manipulation remain persistent problems that organizations find difficult to control. This research designs and builds an employee attendance validation system that combines two layers of verification at once: ArcFace-based facial recognition and location validation through the Global Positioning System (GPS). On the biometric side, the system runs three processing stages: face detection on the image, feature extraction with the ArcFace model, and the formation of a 512- dimensional embedding as a unique numerical representation for each individual. Each attendance attempt is then cross-validated with the device's GPS coordinates to ensure it actually originates from a legitimate work area. When both validations pass, the system records attendance to the database and sends a notification as proof. Test results on five subjects in an office environment show that the system can recognize employee faces consistently even when lighting and background change. The GPS-based location validation also works accurately and effectively prevents location manipulation attempts via fake GPS apps. In other words, the combination of ArcFace and GPS proves to be a more secure, accurate, and reliable attendance solution compared to conventional methods. Kata kunci: attendance validation, biometrics, facial recognition, ArcFace, GPS, digital attendance. Sistem presensi konvensional masih menyimpan banyak celah keamanan. Praktik titip absen, kecurangan lokasi, dan manipulasi jam masuk kerap menjadi masalah yang sulit dikendalikan organisasi. Penelitian ini merancang dan membangun sistem validasi kehadiran pegawai yang memadukan dua lapis verifikasi sekaligus: pengenalan wajah berbasis ArcFace dan validasi posisi melalui Global Positioning System (GPS). Pada sisi biometrik, sistem menjalankan tiga tahap pemrosesan, yaitu deteksi wajah pada citra, ekstraksi fitur dengan model ArcFace, dan pembentukan embedding 512 dimensi sebagai representasi numerik yang khas bagi tiap individu. Setiap upaya presensi kemudian divalidasi secara silang dengan koordinat GPS perangkat agar dapat dipastikan benar-benar berasal dari area kerja yang sah. Jika kedua validasi lolos, sistem mencatat kehadiran ke basis data dan mengirim notifikasi sebagai bukti. Hasil pengujian pada lima subjek di lingkungan kantor menunjukkan sistem dapat mengenali wajah pegawai secara konsisten meski pencahayaan dan latar berubah�ubah. Validasi lokasi berbasis GPS juga bekerja akurat dan efektif mencegah upaya manipulasi posisi melalui aplikasi fake GPS. Dengan kata lain, perpaduan ArcFace dan GPS terbukti menjadi solusi presensi yang lebih aman, akurat, serta andal dibanding metode konvensional. Kata kunci: validasi kehadiran, biometrik, pengenalan wajah, ArcFace, GPS, presensi digital.
Actions (login required)
 |
View Item |
