Smart Learning Library

  • Beranda
  • Informasi
  • News
  • Bantuan
  • Pustakawan
  • Area Anggota
  • Pilih Bahasa :
    Arabic Bengali Brazilian Portuguese English Espanol German Indonesia Bahasa Jepang Melayu Persia Russian Thai Turkish Urdu

Search by:

All Author Subject ISBN/ISSN Advanced Search

Last search:

{{tmpObj[k].text}}
Image of PENCARIAN BERBASIS WAJAH PADA EVENT LARI MENGGUNAKAN MODEL BLAZEFACE DAN METODE DEEP METRIC LEARNING

SKRIPSI IF

PENCARIAN BERBASIS WAJAH PADA EVENT LARI MENGGUNAKAN MODEL BLAZEFACE DAN METODE DEEP METRIC LEARNING

AFIFAH, MUTIARA NURAZMI - Personal Name;

ABSTRAK

Pencarian foto peserta pada dokumentasi running event umumnya masih
dilakukan secara manual sehingga membutuhkan waktu yang lama, terutama ketika
jumlah foto yang dihasilkan mencapai ribuan. Penelitian ini bertujuan untuk
mengembangkan sistem pencarian berbasis wajah pada dokumentasi running event
menggunakan model BlazeFace untuk deteksi wajah dan metode Deep Metric
Learning untuk ekstraksi fitur embedding. Sistem dibangun menggunakan framework
Django sebagai backend, MongoDB sebagai database, dan Redis sebagai lapisan
cache untuk meningkatkan kecepatan respons pencarian.
Pengujian dilakukan menggunakan 1.211 foto dokumentasi dengan empat
subjek uji pada beberapa nilai threshold (0,10; 0,15; 0,20; 0,25; 0,30; 0,35; 0,40; 0,45;
0,50; dan 0,60). Hasil penelitian menunjukkan bahwa sistem berhasil
diimplementasikan dan seluruh fungsi utama berjalan sesuai dengan kebutuhan.
Berdasarkan hasil pengujian, nilai threshold 0.35 dipilih sebagai nilai yang digunakan
pada sistem karena memberikan keseimbangan yang lebih baik antara jumlah foto
yang berhasil ditemukan dan tingkat kesalahan (false positive). Selain itu, hasil
eksperimen cropping manual menunjukkan bahwa kualitas pemotongan wajah
memengaruhi nilai cosine distance sehingga berpengaruh terhadap hasil pencarian.
Implementasi Redis sebagai mekanisme caching juga berhasil meningkatkan efisiensi
pencarian berulang dengan mempercepat waktu respons sistem.

Kata Kunci: Face recognition, BlazeFace, Deep Metric Learning, Running event,
Cosine Distance

vi

ABSTRACT

Searching for participants' photos in running event documentation is generally
performed manually, making the process time-consuming, particularly when
thousands of photos are generated. This study aims to develop a face-based search
system for running event documentation using the BlazeFace model for face detection
and the Deep Metric Learning method for facial embedding extraction. The system
was developed using the Django framework as the backend, MongoDB as the
database, and Redis as a caching layer to improve search response time.
The system was evaluated using 1,211 documentation photos involving four
test subjects with several threshold values (0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40,
0.45, 0.50, and 0.60). The results show that the proposed system was successfully
implemented and that all core functionalities operated as intended. Based on the
experimental results, a threshold value of 0.35 was selected because it provided a
better balance between the number of retrieved photos and the false positive rate. In
addition, manual face-cropping experiments demonstrated that cropping quality
affected the cosine distance values, which consequently influenced the search results.
Furthermore, the implementation of Redis as a caching mechanism improved the
efficiency of repeated search operations by reducing the system response time.

Keywords: Face recognition, BlazeFace, Deep Metric Learning, Running


Ketersediaan
S260908002607.2 AFI pPerpustakaan STMIK AMIKBANDUNGTersedia
Informasi Detil
Judul Seri
-
No. Panggil
607.2 AFI p
Penerbit
: ., 2026
Deskripsi Fisik
-
Bahasa
Indonesia
ISBN/ISSN
-
Klasifikasi
NONE
Tipe Isi
-
Tipe Media
-
Tipe Pembawa
-
Edisi
-
Subyek
SKRIPSI IF
SKRIPSI
Info Detil Spesifik
-
Pernyataan Tanggungjawab
-
Versi lain/terkait

Tidak tersedia versi lain

Lampiran Berkas
Komentar

You must be logged in to post a comment

Smart Learning Library
  • Information
  • Services
  • Librarian
  • Member Area

About Us

As a complete Library Management System, SLiMS (Senayan Library Management System) has many features that will help libraries and librarians to do their job easily and quickly. Follow this link to show some features provided by SLiMS.

Search

start it by typing one or more keywords for title, author or subject

Keep SLiMS Alive Want to Contribute?

© 2026 — Senayan Developer Community

Powered by SLiMS
Select the topic you are interested in
  • Computer science, information & general works
  • Philosophy & psychology
  • Religion
  • Social sciences
  • Language
  • Pure Science
  • Applied sciences
  • Arts & recreation
  • Literature
  • History & geography
Advanced Search