Aplikasi Learning Vector Quantization Dan Principal Component Analysis Pada Pengenalan Individu Melalui Identifikasi Iris Mata

Dewy, Melynda Sylvia (2017) Aplikasi Learning Vector Quantization Dan Principal Component Analysis Pada Pengenalan Individu Melalui Identifikasi Iris Mata. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
1213100070-Undergraduate_Theses.pdf - Published Version

Download (2MB) | Preview

Abstract

Biometrika merupakan cabang matematika terapan yang memanfaatkan karakteristik biologis yang bersifat unik yang dimiliki oleh individu untuk proses identifikasi. Karakteristik unik yang ada pada tubuh individu yang dapat digunakan sebagai biometrika antara lain, seperti wajah, telapak tangan, sidik jari, suara, dan iris mata. Dalam penelitian ini dibuat sistem pengenalan individu melalui identifikasi iris mata menggunakan metode Principal Component Analysis (PCA) untuk mendapatkan ciri dari citra iris mata dan Learning Vector Quantization (LVQ) digunakan untuk proses pelatihan dan pengujian pada tahap klasifikasi. Data yang digunakan dalam penelitian ini adalah data citra mata dari 108 mata berbeda yang diambil dari CASIA database versi 1.0. Uji coba dilakukan pada 30, 50, 80 dan 108 mata yang menghasilkan akurasi sebesar 93.518% pada 108 individu dengan pengambilan 100 vektor eigen, learning rate=0.1, epoch=100 dan penurunan learning rate=0.1. ================================================================= Biometrics is a branch of applied mathematics that utilizes unique biological characteristics possessed by individuals for the identification process. The unique characteristics that exist on the individual body that can be used as biometrics include, such as face, palms, fingerprints, speech, and iris. In this study an individual recognition system was developed through the identification of the iris using Principal Component Analysis (PCA) method to obtain the features of the iris image and Learning Vector Quantization (LVQ) was used for the training and testing process at the classification stage. The data used in this research is eye image data from 108 different eyes taken from CASIA database version 1.0. Trials were performed on 30, 50, 80 and 108 eyes which resulted in an accuracy of 93.518% in 108 individuals with 75 eigen vectors, learning rate = 0.1, epoch = 100 and decreased learning rate = 0.1.

Item Type: Thesis (Undergraduate)
Additional Information: RSMa 006.248 Dew a
Uncontrolled Keywords: biometrika, iris mata, Principal Component Analysis, Learning Vector Quantization
Subjects: Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics)
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Melynda Sylvia Dewy
Date Deposited: 19 Oct 2017 02:46
Last Modified: 06 Mar 2019 01:48
URI: http://repository.its.ac.id/id/eprint/47592

Actions (login required)

View Item View Item