EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Undergraduate Theses » Sistem Informasi - S1
Posted by dewi007 at 13/05/2009 15:07:30  •  10878 Views


IMPLEMENTATION OF MULTIPLE DISCRIMINANT ANALYSIS USING FUKUNAGA KOONTZ TRANSFORM

Author :
Mujahidillah, Muhammad 




ABSTRAK

Analisa Diskriminan Linier merupakan salah satu metode yang sering digunakan dan dikembangkan pada bidang pengenalan pola. Metode ini mencoba menemukan subspace optimal dengan memaksimalkan Fisher Criterion. Penerapan pengenalan pola pada data berdimensi tinggi dan jumlah sample training yang sedikit menyebabkan matriks sebaran within-class bersifat singular. Pada Tugas Akhir ini akan dikembangkan metode Analisa Diskriminan Linier dengan pendekatan Transformasi Fukunaga Koontz untuk memenuhi kebutuhan matriks sebaran withinclass yang bersifat nonsingular. Berdasarkan Transformaasi Fukunaga Koontz seluruh space data didekomposisi ke dalam empat subspace dengan kemampuan diskriminan yang berbeda-beda diukur dari rasio eigenvalue. Fisher Criterion maksimal dapat diketahui dengan menghubungkan rasio eigenvalue dan generalized eigenvalue. Selanjutnya akan diperkenalkan metode baru bernama Analisa Diskriminan Majemuk dengan mentransformasikan data menjadi intraclass dan extraclass dan memaksimalkan jarak Bhattacharyya. Metode ini lebih efisien karena bekerja meskipun matriks sebaran within-class singular dan matriks sebaran between-class bernilai nol.


ABSTRACT

Linear Discriminant Analysis is commonly used and studied in pattern recognition.It try to find optimal subspace by maximizing Fisher Criterion. Implementation of pattern recognition in high dimensional data and few training samples made scatter matrix within-class singular. In this paper Linear Discriminant Analysis were implemented using Fukunaga Koontz Transform to circumvent the requirement of nonsingularity scatter matrix within class. According to Fukunaga Koontz Transform whole data space will be decomposed into four subspaces with different discriminabilities measured by eigenvalue ratio. Maximized Fisher Criterion found by connecting the eigenvalue ratio and generalized eigenvalue. Next a new method called Multiple Discriminant Analysis will be proposed by transforming the data into intraclass and extraclass followed by maximizing the Bhattacharyya distance. This method is more efficient because it worked even if scatter matrix within-class singular and scatter matrix between-class is zero.



Keywordsklasifikasi pola; analisa diskriminan linier; analisa diskriminan majemuk; transformasi fukunaga koontz
 
Subject:  analisis diskriminan
Contributor
  1. Wiwik Anggraeni, S.Si, M.Kom
    Rully Soelaiman, S.Kom, M.Kom
Date Create: 13/05/2009
Type: Text
Format: pdf.
Language: Indonesian
Identifier: ITS-Undergraduate-3100009034098
Collection ID: 3100009034098
Call Number: RSSI 519.535 Muj p


Source
Undergraduate theses of Information System Engineering, RSSI 519.535 Muj p, 2008

Coverage
ITS Community Only

Rights
Copyright @2008 by ITS Library. This publication is protected by copyright and permission should be obtained from the ITS Library prior to any prohibited reproduction, storage in a retrievel system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. For information regarding permission(s), write to ITS Library




[ Download - Summary ]

ITS-Undergraduate-3100009034098-4113.pdf




 Similar Document...




! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 10.199.6.2
using CCBot/2.0 (https://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2021 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan