Perancangan dan Implementasi Face Recognition Management System untuk Identifikasi Otomatis Karyawan

Juan, Adnan Abdullah (2025) Perancangan dan Implementasi Face Recognition Management System untuk Identifikasi Otomatis Karyawan. Project Report. [s.n], [s.l.]. (Unpublished)

Text 5025221155-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Sistem keamanan di lingkungan industri modern membutuhkan solusi identifikasi yang cepat dan akurat untuk menggantikan metode konvensional yang rentan kehilangan dan pemalsuan. Kerja praktik ini mengembangkan Face Recognition Management System (FRMS) untuk identifikasi karyawan otomatis di PT XLSMART Telecom Sejahtera Tbk menggunakan arsitektur client-server yang mengintegrasikan modul AI berbasis MTCNN dan FaceNet untuk deteksi dan ekstraksi embedding wajah, backend FastAPI untuk logika bisnis dan layanan API, serta frontend Node-RED & Vue.js untuk manajemen data dan monitoring real-time. Sistem dirancang dengan penyimpanan terdistribusi yang mengombinasikan database SQLAlchemy untuk metadata, file system untuk gambar, dan file PKL untuk embedding. Identifikasi dilakukan melalui perhitungan cosine similarity antara embedding wajah terdeteksi dengan database tersimpan. Hasil pengujian menunjukkan sistem mampu melakukan identifikasi real-time dengan akurasi melampaui target minimal pada kondisi pencahayaan normal, meningkatkan efisiensi operasional dan keamanan melalui otomasi proses identifikasi berbasis biometrik.
==================================================================================================================================
Security systems in modern industrial environments require fast and accurate identification solutions to replace conventional methods that are prone to loss and forgery. This internship project develops a Face Recognition Management System (FRMS) for automated employee identification at PT XLSMART Telecom Sejahtera Tbk using a client–server architecture. The system integrates AI-based modules employing MTCNN for face detection and FaceNet for facial embedding extraction, a FastAPI backend for business logic and API services, and a Node-RED & Vue.js frontend for data management and real-time monitoring. The system is designed with a distributed storage architecture, combining SQLAlchemy-based databases for metadata, a file system for image storage, and PKL files for facial embeddings. Identification is performed by calculating cosine similarity between detected facial embeddings and those stored in the database. Experimental results show that the system is capable of performing real-time identification with accuracy exceeding the predefined minimum target under normal lighting conditions, thereby improving operational efficiency and security through automated biometric-based identification

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Face recognition, FaceNet, MTCNN FastAPI, Node-RED, Vuejs, Pendeteksi wajah, FaceNet, MTCNN, FastAPI, Node-RED, Vuejs.
Subjects:	T Technology > T Technology (General) > T58.6 Management information systems T Technology > T Technology (General) > T58.64 Information resources management
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Adnan Abdullah Juan
Date Deposited:	23 Dec 2025 05:36
Last Modified:	23 Dec 2025 05:36
URI:	http://repository.its.ac.id/id/eprint/129128

Actions (login required)

View Item