Alkahfi, Muhammad Maulana (2021) Implementasi Metode Viola-Jones Untuk Pengenalan Wajah Pada Sistem Absensi New Normal. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
10311710000072-Undergraduate_Thesis.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Sistem Absensi dengan sidik jari menjadi tidak relevan digunakan karena global pandemik yaitu COVID-19. PT. Galena Perkasa yang punya banyak pegawai masuk dan keluar setiap harinya menjadi rentan sebagai cluster penyebaran. sehingga dibutuhkan absensi dengan minim kontak fisik dan juga penerapan protokol kesehatan sebagai upaya pencegahan penyebaran.Sistem nantinya akan terbagi menjadi 3 bagian yaitu deteksi temperatur tubuh, pengenalan wajah dan juga pendeteksian masker.
Penelitian berfokus pada pengenalan wajah dengan metode Viola-Jones. Viola Jones adalah sebuah kerangka kerja yang terdiri dari haar-like feature untuk mencari fitur fitur pada wajah, Adabost training untuk mempercepat proses dengan training sebelumnya, dan Haar-Feature Classifier dan juga Cascade untuk identifikasi wajah kemudian dilakukan pengenalan dengan menghitung jarak antar titik pada wajah.
Hasil pengujian sistem menunjukkan hasil cukup baik dengan dengan keakuratan 92% pada pengujian ekspresi, 95 % pada pengujian cahaya dan 96% pada pengujian jarak. Dengan waktu rata rata 25 sekon
======================================================================================================
The attendance system with fingerprints has become irrelevant due to the global pandemic, namely COVID-19. PT. Galena Perkasa, which has many employees entering and leaving every day, becomes vulnerable as a cluster spread. so attendance is needed with minimal physical contact and also the application of health protocols as an effort to prevent the spread. The system will later be divided into 3 parts, namely body temperature detection, face recognition and also mask detection.
The research will focus on facial recognition using the Viola-Jones method. Viola Jones is a framework consisting of haar-like features to search for facial features, Adabost training to speed up the process with previous training, and Haar-Feature Classifier and Cascade to determine whether it is a face or not to calculate the distance from existing points. on the face.
The results of the system test showed quite good results with an accuracy of 92% on the expression test, 95% on the light test and 96% on the distance test. With an average time of 25 seconds.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | COVID-19, Pengenalan Wajah, Sistem Absensi, Attendance Systems, Covid-19, Face Recognition |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.B56 Biometric identification T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.P3 Pattern recognition systems |
| Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
| Depositing User: | Muhammad Maulana Alkahfi |
| Date Deposited: | 30 Aug 2021 04:46 |
| Last Modified: | 21 Oct 2024 05:13 |
| URI: | http://repository.its.ac.id/id/eprint/90532 |
Actions (login required)
 |
View Item |