IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2007-03-15 10:30
Additional Optimization of Feature Extraction for Pattern Classification by Utilizing Multilayer Perceptron
Norikazu Tonogai, Shin Ishii, Tomohiro Shibata (NAST)
Abstract (in Japanese) (See Japanese page) 
(in English) The demand of real-time pattern classification of high-dimensional
data with highly restricted computational power isincreasing. Possible applications would be face recognition in a car or by a mobile phone.
Under these situations, reduction of computing time by mean of feature extraction which projects high-dimensional data into a low-dimensional space. One of the most popular feature extraction method is principle component analysis (PCA). PCA finds an optimal feature space that
maximizes the mutual information between the reduced data and the
original data, but this does not mean the found space is also optimal
for pattern classification, in general. The aim of this study is to propose an approach to acquire such a feature extractor that is suited
for pattern classification. We employ a three-layer perceptron for this aim. Because learning of a high-dimensional feature extractor usually
finds a local minimum, we propose to use PCA to find a good initial parameter set for the weights between the input and the intermediate layers. Computer experiments using two face databases popular in the research area of face recognition demonstrates the plausibility of our approach, even when it is compared to support vector machine.
Keyword (in Japanese) (See Japanese page) 
(in English) pattern classification / feature extraction / principal component analysis / multilayer perceptron / / / /  
Reference Info. IEICE Tech. Rep., vol. 106, no. 589, NC2006-172, pp. 111-116, March 2007.
Paper # NC2006-172 
Date of Issue 2007-03-08 (NC) 
ISSN Print edition: ISSN 0913-5685
Download PDF

Conference Information
Committee NC  
Conference Date 2007-03-14 - 2007-03-16 
Place (in Japanese) (See Japanese page) 
Place (in English) Tamagawa University 
Topics (in Japanese) (See Japanese page) 
Topics (in English) General 
Paper Information
Registration To NC 
Conference Code 2007-03-NC 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Additional Optimization of Feature Extraction for Pattern Classification by Utilizing Multilayer Perceptron 
Sub Title (in English)  
Keyword(1) pattern classification  
Keyword(2) feature extraction  
Keyword(3) principal component analysis  
Keyword(4) multilayer perceptron  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Norikazu Tonogai  
1st Author's Affiliation Nara Institute of Science and Technology (NAST)
2nd Author's Name Shin Ishii  
2nd Author's Affiliation Nara Institute of Science and Technology (NAST)
3rd Author's Name Tomohiro Shibata  
3rd Author's Affiliation Nara Institute of Science and Technology (NAST)
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2007-03-15 10:30:00 
Presentation Time 20 minutes 
Registration for NC 
Paper # NC2006-172 
Volume (vol) vol.106 
Number (no) no.589 
Page pp.111-116 
#Pages
Date of Issue 2007-03-08 (NC) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan