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 2008-03-06 14:45
Introduce Batch-Learning Algorithm into the Contour Extraction Method DCDAM using Self-Organizing Map
Fukuya Namba (Tottori Univ. of Environmental Studies), Takuya Ueta (GALAXY Inc.), Yasuaki Sumi (Tottori Univ. of Environmental Studies), Noboru Yabuki (Tsuyama National College of Tech.), Takao Tsukutani (Matsue National College of Tech.) CAS2007-121 SIP2007-196 CS2007-86
Abstract (in Japanese) (See Japanese page) 
(in English) In this paper, we proposed the new contour line extraction method BL-DCDAM introduced the Batch-Learning SOM algorithm into DCDAM we have proposed. DCDAM can extract the contour line from the gray scale image by using self-organizing map (Self-Organizing Map: SOM). DCDAM training has adopted Sequential-Learning SOM algorithm. So in a case of polygonal target object it can extract contour line in some degree it is not so good result than we have expected depending on the sequence of presenting the input vector at the training process. To solve this problem, we have examined Batch-Learning SOM algorithm as a method which does not depend on the sequence of presenting the input vector. When the reference vector is updated, the trainings can be achieved both by updating the reference vector on the competitive layer and by getting the mean vector of the input vector group that exists in the reference vector included in neighborhood in the Voronoi region at Batch-Learning SOM algorithm. Then, in our research to introduce Batch-Learning SOM algorithm into the learning algorithm of DCDAM, we have developed BL-DCDAM, that is, the new method to adjust the degree of the update by putting the weight to each input vector (coordinate value of the pixel) according to the weighting function when this mean vector is calculated. As one example, we have experiments on the road sign as an extraction target. We have obtained an especially excellent result about a polygonal target object (road sign) compared with the Sequential-Learning SOM algorithm.
Keyword (in Japanese) (See Japanese page) 
(in English) Contour Extraction / DCDAM / Self-Organizing Map(SOM) / Batch-Learning SOM Algorithm / / / /  
Reference Info. IEICE Tech. Rep., vol. 107, no. 528, SIP2007-196, pp. 75-80, March 2008.
Paper # SIP2007-196 
Date of Issue 2008-02-28 (CAS, SIP, CS) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF CAS2007-121 SIP2007-196 CS2007-86

Conference Information
Committee CS SIP CAS  
Conference Date 2008-03-06 - 2008-03-07 
Place (in Japanese) (See Japanese page) 
Place (in English) Yamaguchi University 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Network processors, signal processing for communications, wireless LAN/PAN, etc. 
Paper Information
Registration To SIP 
Conference Code 2008-03-CS-SIP-CAS 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Introduce Batch-Learning Algorithm into the Contour Extraction Method DCDAM using Self-Organizing Map 
Sub Title (in English)  
Keyword(1) Contour Extraction  
Keyword(2) DCDAM  
Keyword(3) Self-Organizing Map(SOM)  
Keyword(4) Batch-Learning SOM Algorithm  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Fukuya Namba  
1st Author's Affiliation Tottori University of Environmental Studies (Tottori Univ. of Environmental Studies)
2nd Author's Name Takuya Ueta  
2nd Author's Affiliation GALAXY Inc. (GALAXY Inc.)
3rd Author's Name Yasuaki Sumi  
3rd Author's Affiliation Tottori University of Environmental Studies (Tottori Univ. of Environmental Studies)
4th Author's Name Noboru Yabuki  
4th Author's Affiliation Tsuyama National College of Technology (Tsuyama National College of Tech.)
5th Author's Name Takao Tsukutani  
5th Author's Affiliation Matsue National College of Technology (Matsue National College of Tech.)
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 2008-03-06 14:45:00 
Presentation Time 25 minutes 
Registration for SIP 
Paper # CAS2007-121, SIP2007-196, CS2007-86 
Volume (vol) vol.107 
Number (no) no.526(CAS), no.528(SIP), no.530(CS) 
Page pp.75-80 
#Pages
Date of Issue 2008-02-28 (CAS, SIP, CS) 


[Return to Top Page]

[Return to IEICE Web Page]


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