Pengenalan Individu Melalui Identifikasi Wajah Berbasis Metode You Only Look Once (Yolov5).

Diaz, Athaya Aufa (2022) Pengenalan Individu Melalui Identifikasi Wajah Berbasis Metode You Only Look Once (Yolov5). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111840000113-Undergraduate-Thesis.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Biometrika merupakan cabang matematika terapan untuk mengenali individu yang memanfaatkan identifikasi karakteristik atau ciri khusus yang dimilikinya. Saat ini sistem pengenalan individu melalui identifikasi wajah menjadi salah satu teknologi biometrika yang sedang populer. Di balik popularitasnya, sistem pengenalan wajah memiliki kerumitan tersendiri dalam proses identifikasinya. Seiring berkembangnya zaman, sistem dituntut untuk mampu mengenali wajah secara real time. You Only Look Once (YOLO) merupakan salah satu sistem deteksi objek yang terkenal dengan kemampuannya dalam mendeteksi objek secara real time. Terlebih lagi, YOLO versi kelima atau YOLOv5 tidak memerlukan memori yang besar. Penelitian ini bertujuan untuk mengaplikasikan metode YOLOv5 dalam pengenalan individu melalui identifikasi wajah dan menganalisis kinerjanya. Metode YOLOv5 dipilih karena kemampuannya dalam mendeteksi objek dengan cepat. Data yang digunakan dalam penelitian ini berupa citra digital video dari beberapa mahasiswa Institut Teknologi Sepuluh Nopember (ITS) yang diambil menggunakan kamera CCTV. Tahapan penelitian meliputi pengumpulan data, pembagian data (data latih dan data uji), pra-pemrosesan (akuisisi video, ekstraksi frame, dan anotasi data), pelatihan model dengan hyperparameter tertentu, dan pengujian model. Hasil penelitian menunjukkan bahwa penerapan metode YOLOv5 efektif dalam mengenali individu melalui identifikasi wajah dengan waktu pemrosesan yang cepat. Model yang dihasilkan mampu memberikan akurasi yang baik dalam pengenalan individu pada berbagai kondisi. Penelitian ini memberikan kontribusi dalam pengembangan sistem identifikasi wajah berbasis deteksi objek yang cepat dan akurat.
==================================================================================================================================
Biometrics is a branch of applied mathematics for identifying individuals that utilizes the identification of characteristics or specific traits they possess. Currently, individual recognition systems through face identification have become one of the most popular biometric technologies. Behind its popularity, face recognition systems have their own complexity in the identification process. As time progresses, systems are required to be able to recognize faces in real time. You Only Look Once (YOLO) is one of the object detection systems known for its ability to detect objects in real time. Furthermore, the fifth version of YOLO or YOLOv5 does not require a large amount of memory. This study aims to apply the YOLOv5 method in individual recognition through face identification and analyze its performance. The YOLOv5 method was chosen due to its ability to detect objects quickly. The data used in this study consists of digital video images of several students from the Institut Teknologi Sepuluh Nopember (ITS) taken using a CCTV camera. The research stages include data collection, data splitting (training set and testing set), pre-processing (video acquisition, frame extraction, and data annotation), model training with specific hyperparameters, and model testing. The research results indicate that the application of the YOLOv5 method is effective in recognizing individuals through face identification with fast processing times. The resulting model is capable of providing good accuracy in individual recognition under various conditions. This research contributes to the development of face identification systems based on fast and accurate object detection.

Item Type:	Thesis (Other)
Additional Information:	RSMa 006.42 Dia p-1 2022
Uncontrolled Keywords:	Biometrika. Pengenalan wajah. YOLOv5. Biometrics. Face recognition. YOLOv5. Biometrics, Individual Recognition, Face Identification, YOLOv5.
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	10 Jun 2026 01:02
Last Modified:	10 Jun 2026 01:02
URI:	http://repository.its.ac.id/id/eprint/133661

Actions (login required)

View Item