Abdul Malik, Adam (2023) IDENTIFIKASI WAJAH MANUSIA MENGGUNAKAN YOLO FRAMEWORKS DENGAN METODE SCALE MODIFIER SEBAGAI PREPROCESSING SECARA REAL TIME. S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.

	Text Adam Abdul Malik_3332160050_Fulltext.pdf Download (3MB)
	Text Adam Abdul Malik_3332160050_01.pdf Download (372kB)
	Text Adam Abdul Malik_3332160050_02.pdf Download (361kB)
	Text Adam Abdul Malik_3332160050_03.pdf Download (328kB)
	Text Adam Abdul Malik_3332160050_04.pdf Download (990kB)
	Text Adam Abdul Malik_3332160050_05.pdf Download (67kB)
	Text Adam Abdul Malik_3332160050_Ref.pdf Download (144kB)
	Text Adam Abdul Malik_3332160050_Lamp.pdf Download (1MB)
	Text Adam Abdul Malik_3332160050_CP.pdf Download (28MB)

Abstract

ABSTRACT Adam Abdul Malik Electrical Engineering Human Face Identification Using Yolo Frameworks with Scale Modifier Method as Real Time Preprocessing . Technological developments have increased very rapidly in today's era, this study designed a system that uses the You only look once (YOLO) method for facial recognition. YOLO models must be trained in human facial identification. The training process is carried out in Google Colaboratory. The results of the training process are then optimized so that it will reduce the file size. After the YOLOv4.weight model is created, a program for real-time identification of human faces can be created using You only look once (YOLO). In face detection it has an accuracy rate of 94.193% with the camera angle in the front position with a total of 20 face data and the FPS results obtained on average 2,8 FPS. Keywords: Human Face Identification, You Only Look Once (YOLO), Python, GoogleColab.

Item Type:	Thesis (S1)
Contributors:	Contribution Contributors NIP/NIM Thesis advisor FAHRIZAL, RIAN 197510262005011001 Thesis advisor ADIPURA WICAKSANA, CAKRA 199006282019031010
Additional Information:	ABSTRAK Adam Abdul Malik Teknik Elektro Identifikasi Wajah Manusia Menggunakan Yolo Frameworks Dengan Metode Scale Modifier Sebagai Preprocessing Secara Real Time Perkembangan teknologi meningkat sangat pesat di era sekarang ini, penelitian ini merancang sebuah sistem yang menggunakan metode You only look once (YOLO) untuk pengenalan wajah. Model YOLO harus melewati proses training untuk identifikasi wajah manusia. Proses training dilakukan di Google Colaboratory. Hasil dari proses training kemudian di optimalisasi sehingga akan memperkecil ukuran file, Setelah model YOLOv4.weight dibuat maka dapat membuat program untuk identifikasi wajah manusia secara realtime menggunakan You only look once (YOLO). Dalam deteksi wajah memiliki tingkat akurasi sebasar 94,193% dengan sudut kamera berada di posisi depan dengan total data wajah sebanyak 20 dan hasil FPS yang didapat rata-rata 2,8 FPS. Kata Kunci: Identifikasi wajah manusia, You Only Look Once (YOLO), Python, Google Colab.
Subjects:	Communication > Science Journalism
Divisions:	03-Fakultas Teknik 03-Fakultas Teknik > 20201-Jurusan Teknik Elektro
Depositing User:	Mr Adam Abdul Malik
Date Deposited:	16 Oct 2023 08:23
Last Modified:	16 Oct 2023 08:23
URI:	http://eprints.untirta.ac.id/id/eprint/30442

Actions (login required)

View Item