Perancangan Face Recognition Pada Sistem Presensi Mahasiwa Menggunakan Metode YOLOv11

Ozza, Jardine Fakhri Ozza (2026) Perancangan Face Recognition Pada Sistem Presensi Mahasiwa Menggunakan Metode YOLOv11. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040211103-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Presensi mahasiswa merupakan salah satu komponen penting dalam proses pembelajaran karena berfungsi sebagai indikator kedisiplinan dan keaktifan mahasiswa. Namun, sistem presensi manual maupun berbasis fingerprint masih memiliki berbagai keterbatasan, seperti membutuhkan waktu yang relatif lama, berpotensi menimbulkan antrean, rentan terhadap kecurangan berupa titip absen, serta kurang efisien dalam proses pencatatan kehadiran. Perkembangan teknologi computer vision dan deep learning memberikan peluang untuk mengembangkan sistem presensi yang lebih cepat, akurat, dan otomatis melalui penerapan teknologi face recognition. Penelitian ini bertujuan untuk merancang sistem presensi mahasiswa berbasis pengenalan wajah menggunakan metode YOLOv11 yang mampu melakukan deteksi dan identifikasi wajah secara real-time. Sistem dirancang dengan tahapan pengambilan citra wajah, pre-processing data, pelatihan model YOLOv11, proses deteksi dan pengenalan wajah, serta pencatatan hasil presensi secara otomatis ke dalam basis data. Implementasi sistem direncanakan pada Laboratorium Eldrive AA104 sebagai media uji penerapan. Penggunaan YOLOv11 dipilih karena memiliki keunggulan dalam kecepatan deteksi, akurasi yang tinggi, serta kemampuan mengenali banyak objek dalam satu citra secara simultan. Hasil evaluasi performa model menunjukkan bahwa metode YOLOv11 memiliki kemampuan yang sangat baik dalam melakukan deteksi dan pengenalan wajah. Model menghasilkan nilai mAP@0.5 sebesar 98,9%, F1-Score sebesar 99%, serta akurasi pengujian mencapai 95,92% berdasarkan hasil evaluasi confusion matrix. Nilai precision, recall, dan F1-score pada masing-masing kelas juga menunjukkan performa yang tinggi sehingga model dinilai stabil dan andal.
======================================================================================================================================
Student attendance is a key component of the learning process, as it serves as an indicator of student discipline and engagement. However, both manual and fingerprint-based attendance systems still have various limitations, such as being relatively time-consuming, potentially causing lines, being susceptible to fraud such as having someone else sign in on one’s behalf, and being inefficient in the attendance recording process. Advances in computer vision and deep learning technologies offer the opportunity to develop a faster, more accurate, and automated attendance system through the application of face recognition technology. This study aims to design a student attendance system based on face recognition using the YOLOv11 method, which is capable of detecting and identifying faces in real time. The system is designed with stages for capturing facial images, data preprocessing, training the YOLOv11 model, the face detection and recognition process, and automatically recording attendance results into a database. The system is planned to be implemented in the Eldrive AA104 Laboratory as a testing environment. YOLOv11 was chosen for its advantages in detection speed, high accuracy, and the ability to simultaneously recognize multiple objects in a single image. The model performance evaluation results show that the YOLOv11 method performs exceptionally well in face detection and recognition. The model achieved a mAP@0.5 of 98.9%, an F1-Score of 99%, and a test accuracy of 95.92% based on the confusion matrix evaluation results. The precision, recall, and F1-score values for each class also demonstrated high performance, indicating that the model is stable and reliable.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	face recognition, Presensi Mahasiswa, YOLOv11, Computer Vision, Deep Learning, Deteksi wajah, Face Recognition, Student Attendance,YOLOv11, Computer Vision, Deep Learning, Face Detection
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1650 Face recognition. Optical pattern recognition.
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Jardine Fakhri Ozza
Date Deposited:	11 Aug 2026 03:07
Last Modified:	11 Aug 2026 03:07
URI:	http://repository.its.ac.id/id/eprint/144290

Actions (login required)

View Item