RANCANG BANGUN PROTOTYPE FACE RECOGNITION BERBASIS YOLO11 DENGAN MENGGUNAKAN RASPBERRY PI

LISTIA SETIAWATI, . (2025) RANCANG BANGUN PROTOTYPE FACE RECOGNITION BERBASIS YOLO11 DENGAN MENGGUNAKAN RASPBERRY PI. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (1MB)
	Text BAB 1.pdf Download (295kB)
	Text BAB 2.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (439kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (281kB) | Request a copy
	Text Daftar Pustaka.pdf Download (314kB)
	Text Lampiran.pdf Restricted to Registered users only Download (788kB) | Request a copy

Abstract

Penelitian ini bertujuan untuk merancang prototype face recognition berbasis YOLO11 dengan menggunakan Raspberry Pi. Perancangan Prototype face recognition menggunakan Raspberry Pi 4B, dengan input Raspberry Pi Camera Module 2 dengan output berupa audio hasil identifikasi identitas berdasarkan face recognition.Pada penelitian ini menggunakan dataset face dengan total 250 foto dengan 10 class atau label, yang artinya prototype ini dapat mengidentifikasi face recognition dari 10 orang. Pembagian dataset adalah 80% (200) gambar wajah untuk training, 8% (20) gambar wajah untuk testing, dan 12% (30) gambar wajah untuk validasi. Berdasarkan hasil pengujian alat pada 10 video bergerak didapatkan nilai akurasi sebesar 95 %, presisi 100% recall 89.6%, dan F1-Score sebesar 94%. Meskipun demikian, performa identifikasi rentan terhadap kondisi pencahayaan backlight, motion blur, dan pose wajah yang terlalu menoleh ekstrem dapat mengurangi akurasi deteksi. Pengujian Task Success Rate untuk mengukur kinerja speaker mencapai 100%, menandakan fungsi notifikasi sistem dalam memberitahukan hasil identifikasi sesuai dengan hasil pembacaan info log, membuktikan kelayakan prototype face recognition berbasis YOLO11 dengan menggunakan Raspberry Pi. ***** This research aims to design and develop a YOLO11-based face recognition prototype using a Raspberry Pi. The increasing need for efficient and deployable personal identification solutions on resource-constrained devices drives the design of the prototype. The system utilizes a Raspberry Pi 4B, with the Raspberry Pi Camera Module 2 as its visual input, and provides audio output based on face identification results.For this study, a face dataset comprising a total of 250 photos across 10 classes (identities) was used, enabling the prototype to identify 10 distinct individuals. The dataset was partitioned with 80% (200 images) for training, 8% (20 images) for testing, and 12% (30 images) for model validation. Experimental results from testing the prototype with 10 moving face videos demonstrated robust performance, achieving an accuracy of 95%, precision of 100%, recall of 89.6%, and an F1-score of 94%. Nevertheless, identification performance was observed to be susceptible to challenging conditions such as strong backlight, motion blur due to rapid movement, and extreme head poses, which could reduce detection accuracy. Furthermore, the Task Success Rate (TSR) test, which measures the speaker's performance in delivering identification notifications, achieved a 100% success rate. This indicates that the system's audio notification function accurately conveyed identification results as per the log information, thereby demonstrating the feasibility and effectiveness of the YOLO11-based face recognition prototype on Raspberry Pi

Item Type:	Thesis (Sarjana)
Additional Information:	1). Rafiuddin Syam, ST., M.Eng, PhD ; 2). Dr. Aodah Diamah, S.T, M.Eng.
Subjects:	Teknologi dan Ilmu Terapan > Teknik Elektronika
Divisions:	FT > S1 Pendidikan Teknik Elektronika
Depositing User:	Listia Setiawati .
Date Deposited:	08 Aug 2025 03:28
Last Modified:	08 Aug 2025 03:28
URI:	http://repository.unj.ac.id/id/eprint/58002

Actions (login required)

View Item