Pengenalan Pola Wajah Untuk Mendukung Absensi Pegawai Menggunakan Metode Local Binary Pattern (LBP)
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Wijayanti, Eka Putri Fajar
Haris, Abdul
Siregar, Riki Ruli Affandi
Haris, Abdul
Siregar, Riki Ruli Affandi
Subject
Teknik Informatika
Datestamp
2022-09-20 07:25:35
Abstract :
The purpose of this study is to recognize facial patterns / face recognition that will be implemented in employee attendances, due to frequent irregularities. In this study using the Local Binary Pattern (LBP) method to face recognition patterns. LBP is a feature extraction method that is quite fast with a low computational process, the way LBP operators work is by finding the middle value of a matrix measuring 3 x 3, by comparing the middle pixel value with the nearest pixel neighbor value on a grayscale image. To measure the degree of similarity, Euclidean Distance is used. From the test results finally it was found that the Local Binary Pattern (LBP) method can be used for face recognition in the attendances system with the accuracy obtained from the validation of Mean Absolute Percentage Error (MAPE) of 85%.
The purpose of this study is to recognize facial patterns / face recognition that will be implemented in employee attendances, due to frequent irregularities. In this study using the Local Binary Pattern (LBP) method to face recognition patterns. LBP is a feature extraction method that is quite fast with a low computational process, the way LBP operators work is by finding the middle value of a matrix measuring 3 x 3, by comparing the middle pixel value with the nearest pixel neighbor value on a grayscale image. To measure the degree of similarity, Euclidean Distance is used. From the test results finally it was found that the Local Binary Pattern (LBP) method can be used for face recognition in the attendances system with the accuracy obtained from the validation of Mean Absolute Percentage Error (MAPE) of 85%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
