Pengenalan Wajah Menggunakan Metode Yolo Pada Sistem Presensi Pegawai

Pambudi, Bintang Ramadhan (2024) Pengenalan Wajah Menggunakan Metode Yolo Pada Sistem Presensi Pegawai. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 10311910000068_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (6MB) | Request a copy

Abstract

Penerapan metode pengenalan wajah berbasis YOLO (You Only Look Once) pada sistem presensi pegawai dapat menjadi alternatif yang efisien dan aman untuk menggantikan metode presensi konvensional, seperti fingerprint dan RFID. Penelitian ini bertujuan untuk merancang sistem presensi pegawai berbasis pengenalan wajah, mengkaji akurasi dalam mengenali dan membedakan wajah pegawai, serta mengevaluasi efisiensi waktu dan peningkatan keamanan dalam proses pencatatan kehadiran. Hasil penelitian menunjukkan bahwa sistem berbasis YOLO memiliki rata-rata waktu deteksi 1,15 detik per individu, yang secara signifikan lebih cepat dibandingkan dengan sistem fingerprint (3,43 detik) dan sistem RFID (4,61 detik). Selain itu, akurasi pengenalan wajah mencapai 80%, menjadikan sistem ini sebagai solusi dalam hal efisiensi waktu dan keandalan deteksi. Dengan demikian, penggunaan metode YOLO dapat meningkatkan efisiensi dalam pencatatan presensi pegawai secara real-time, sehingga memberikan nilai tambah yang signifikan dibandingkan metode presensi konvensional.
=================================================================================================================================
The implementation of YOLO (You Only Look Once) face recognition method in employee attendance systems presents an efficient and secure alternative to conventional methods such as fingerprint and RFID. This study aims to design an employee attendance system based on face recognition, evaluate its accuracy in identifying and distinguishing employee faces, and assess its time efficiency and security improvements in the attendance recording process. The results indicate that the YOLO-based system has an average detection time of 1.15 seconds per individual, significantly faster compared to the fingerprint system (3.43 seconds) and the RFID system (4.61 seconds). Additionally, the face recognition accuracy reaches 80%, positioning the system as a solution in terms of time efficiency and detection reliability. Therefore, the use of the YOLO method can enhance real-time attendance recording efficiency, offering significant added value over conventional attendance methods.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Presensi, YOLO, PengenalanWajah, Attendance, FaceRecognition
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.B56 Biometric identification T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7888.3 Digital computers
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Bintang Ramadhan Pambudi
Date Deposited:	27 Sep 2024 01:46
Last Modified:	27 Sep 2024 01:46
URI:	http://repository.its.ac.id/id/eprint/115699

Actions (login required)

View Item