PENGENALAN WAJAH MENGGUNAKAN METODE LBPH DAN EIGENFACE BERBASIS MORFOLOGI MATEMATIKA

WIDYANTORO, FYAN ESTU (2021) PENGENALAN WAJAH MENGGUNAKAN METODE LBPH DAN EIGENFACE BERBASIS MORFOLOGI MATEMATIKA. S2 thesis, Universitas Mercu Buana Jakarta.

[img] Text (HAL COVER)
COVER FYAN ESTU WIDYANTORO.pdf
Download (369kB)
[img] Text (BAB I)
BAB I FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (113kB)
[img] Text (BAB II)
BAB II FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (509kB)
[img] Text (BAB III)
BAB III FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (133kB)
[img] Text (BAB IV)
BAB IV FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (372kB)
[img] Text (BAB V)
BAB V FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (30kB)
[img] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA COVER FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (93kB)
[img] Text (LAMPIRAN)
LAMPIRAN COVER FYAN ESTU WIDYANTORO.pdf
Restricted to Registered users only
Download (108kB)

Abstract

Human facial recognition system is a rapidly growing field. Real-time facial recognition systems using cameras are needed to detect human identity based on their faces. Haar-cascade is used to detect human faces. There are 400 human facial data that will be observed from 40 different people obtained from the AT&T facial database. In each person there are 10 combinations of faces, for example the front view, left side, right side, and others. Histogram equalization and mathematical morphology (opening and closing) are used in the pre-processing stage. After getting the results of pre-processing, the next step is the feature extraction process, in this study will compare two feature extraction methods, namely Local Binary Pattern Histogram (LBPH) and Eigenface. Face recognition process will be carried out in real time and the detection results will be recorded using LBPH and Eigenface. From these two methods, the required processing time and the accuracy of the detection results will be calculated before using mathematical morphology and after using opening and closing mathematical morphology. The highest accuracy result is in pre-processing using Histogram Equalization and Morphology Mathematical Opening and LBPH feature extraction with an accuracy rate of 98.75%. Meanwhile, in the extraction of Eigenface features, the highest accuracy uses pre-processing Histogram Equalization and Mathematical Morphology Opening with an accuracy rate of 71.25%. Keyword: Face Recognition, Histogram Equalization, Mathematical Morphology, Local Binary Pattern Histogram, Eigenface. Sistem pengenalan wajah manusia merupakan bidang yang berkembang pesat. Sistem pengenalan wajah (face recognition) secara real-time menggunakan kamera banyak dibutuhkan untuk mendeteksi identitas manusia berdasarkan wajahnya. Untuk mendeteksi wajah manusia digunakan haar-cascade. Terdapat 400 data wajah manusia yang akan diamati dari 40 orang yang berbeda diperoleh dari basis data wajah AT&T. Pada tahap pre-processing akan dilakukan histogram equalization dan mathematical morphology (opening dan closing). Setelah didapatkan hasil pre-processing dilakukan proses ekstraksi fitur, pada penelitian ini akan membandingkan dua metode ekstraksi fitur yaitu Local Binary Pattern Histogram (LBPH) dan Eigenface. Kemudian akan dilakukan proses pengenalan wajah secara real time dan pencatatan hasil deteksi menggunakan LBPH dan Eigenface. Dari dua metode tersebut akan dihitung lama proses yang dibutuhkan dan tingkat akurasi hasil deteksi sebelum menggunakan morfologi matematika dan setelah menggunakan morfologi matematika opening dan closing. Hasil akurasi paling tinggi menggunakan pre-processing menggunakan Histogram Equalization dan Morfologi Matematika Opening dan ekstraksi fitur LBPH dengan tingkat akurasi sebesar 98.75%. Sedangkan pada ekstraksi fitur Eigenface akurasi tertinggi menggunakan pre-processing Histogram Equalization dan Morfologi Matematika Opening dengan tingkat akurasi sebesar 71.25%. Keyword: Face Recognition, Histogram Equalization, Mathematical Morphology, Local Binary Pattern Histogram, Eigenface

Item Type: Thesis (S2)
NIM/NIDN Creators: 55417120021
Uncontrolled Keywords: Keyword: Face Recognition, Histogram Equalization, Mathematical Morphology, Local Binary Pattern Histogram, Eigenface
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 001 Knowledge/Ilmu Pengetahuan > 001.4 Research; Statistical Methods/Riset; Metode Statistik
000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 001 Knowledge/Ilmu Pengetahuan > 001.4 Research; Statistical Methods/Riset; Metode Statistik > 001.42 Reseach Methods/Metode Riset
000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 001 Knowledge/Ilmu Pengetahuan > 001.4 Research; Statistical Methods/Riset; Metode Statistik > 001.42 Reseach Methods/Metode Riset > 001.422 Statistical Methods/Metode Statistik
Divisions: Pascasarjana > Magister Teknik Elektro
Depositing User: Dede Muksin Lubis
Date Deposited: 09 Oct 2023 02:23
Last Modified: 09 Oct 2023 02:23
URI: http://repository.mercubuana.ac.id/id/eprint/82193

Actions (login required)

View Item View Item
UMB Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.