A new self-organization classification algorithm for remote-sensing images

Souichi Oka, Tomoaki Ogawa, Takayoshi Oda, Yoshiyasu Takefuji

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This paper presents a new self-organization classification algorithm for remote-sensing images. Kohonen and other scholars have proposed self-organization algorithms. Kohonen's model easily converges to the local minimum by tuning the elaborate parameters. In addition to others, S.C. Amatur and Y. Takefuji have also proposed self-organization algorithm model. In their algorithm, the maximum neuron model (winner-take-all neuron model) is used where the parameter-tuning is not needed. The algorithm is able to shorten the computation time without a burden on the parameter-tuning. However, their model has a tendency to converge to the local minimum easily. To remove these obstacles produced by the two algorithms, we have proposed a new self-organization algorithm where these two algorithms are fused such that the advantages of the two algorithms are combined. The number of required neurons is the number of pixels multiplied by the number of clusters. The algorithm is composed of two stages: in the first stage we use the maximum self-organization algorithm until the state of the system converges to the local-minimum, then, the Kohonen self-organization algorithm is used in the last stage in order to improve the solution quality by escaping from the local minimum of the first stage. We have simulated a LANDSAT-TM image data with 500 pixel × 100 pixel image and 8-bit gray scaled. The results justifies all our claims to the proposed algorithm.

Original languageEnglish
Pages (from-to)132-136
Number of pages5
JournalIEICE Transactions on Information and Systems
VolumeE81-D
Issue number1
Publication statusPublished - 1998

Fingerprint

Remote sensing
Neurons
Tuning
Pixels

Keywords

  • Classification
  • Neural network
  • Remote-sensing
  • Self-organization

ASJC Scopus subject areas

  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

A new self-organization classification algorithm for remote-sensing images. / Oka, Souichi; Ogawa, Tomoaki; Oda, Takayoshi; Takefuji, Yoshiyasu.

In: IEICE Transactions on Information and Systems, Vol. E81-D, No. 1, 1998, p. 132-136.

Research output: Contribution to journalArticle

Oka, Souichi ; Ogawa, Tomoaki ; Oda, Takayoshi ; Takefuji, Yoshiyasu. / A new self-organization classification algorithm for remote-sensing images. In: IEICE Transactions on Information and Systems. 1998 ; Vol. E81-D, No. 1. pp. 132-136.
@article{be8f5bf69f374986b119ce953c84f696,
title = "A new self-organization classification algorithm for remote-sensing images",
abstract = "This paper presents a new self-organization classification algorithm for remote-sensing images. Kohonen and other scholars have proposed self-organization algorithms. Kohonen's model easily converges to the local minimum by tuning the elaborate parameters. In addition to others, S.C. Amatur and Y. Takefuji have also proposed self-organization algorithm model. In their algorithm, the maximum neuron model (winner-take-all neuron model) is used where the parameter-tuning is not needed. The algorithm is able to shorten the computation time without a burden on the parameter-tuning. However, their model has a tendency to converge to the local minimum easily. To remove these obstacles produced by the two algorithms, we have proposed a new self-organization algorithm where these two algorithms are fused such that the advantages of the two algorithms are combined. The number of required neurons is the number of pixels multiplied by the number of clusters. The algorithm is composed of two stages: in the first stage we use the maximum self-organization algorithm until the state of the system converges to the local-minimum, then, the Kohonen self-organization algorithm is used in the last stage in order to improve the solution quality by escaping from the local minimum of the first stage. We have simulated a LANDSAT-TM image data with 500 pixel × 100 pixel image and 8-bit gray scaled. The results justifies all our claims to the proposed algorithm.",
keywords = "Classification, Neural network, Remote-sensing, Self-organization",
author = "Souichi Oka and Tomoaki Ogawa and Takayoshi Oda and Yoshiyasu Takefuji",
year = "1998",
language = "English",
volume = "E81-D",
pages = "132--136",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "1",

}

TY - JOUR

T1 - A new self-organization classification algorithm for remote-sensing images

AU - Oka, Souichi

AU - Ogawa, Tomoaki

AU - Oda, Takayoshi

AU - Takefuji, Yoshiyasu

PY - 1998

Y1 - 1998

N2 - This paper presents a new self-organization classification algorithm for remote-sensing images. Kohonen and other scholars have proposed self-organization algorithms. Kohonen's model easily converges to the local minimum by tuning the elaborate parameters. In addition to others, S.C. Amatur and Y. Takefuji have also proposed self-organization algorithm model. In their algorithm, the maximum neuron model (winner-take-all neuron model) is used where the parameter-tuning is not needed. The algorithm is able to shorten the computation time without a burden on the parameter-tuning. However, their model has a tendency to converge to the local minimum easily. To remove these obstacles produced by the two algorithms, we have proposed a new self-organization algorithm where these two algorithms are fused such that the advantages of the two algorithms are combined. The number of required neurons is the number of pixels multiplied by the number of clusters. The algorithm is composed of two stages: in the first stage we use the maximum self-organization algorithm until the state of the system converges to the local-minimum, then, the Kohonen self-organization algorithm is used in the last stage in order to improve the solution quality by escaping from the local minimum of the first stage. We have simulated a LANDSAT-TM image data with 500 pixel × 100 pixel image and 8-bit gray scaled. The results justifies all our claims to the proposed algorithm.

AB - This paper presents a new self-organization classification algorithm for remote-sensing images. Kohonen and other scholars have proposed self-organization algorithms. Kohonen's model easily converges to the local minimum by tuning the elaborate parameters. In addition to others, S.C. Amatur and Y. Takefuji have also proposed self-organization algorithm model. In their algorithm, the maximum neuron model (winner-take-all neuron model) is used where the parameter-tuning is not needed. The algorithm is able to shorten the computation time without a burden on the parameter-tuning. However, their model has a tendency to converge to the local minimum easily. To remove these obstacles produced by the two algorithms, we have proposed a new self-organization algorithm where these two algorithms are fused such that the advantages of the two algorithms are combined. The number of required neurons is the number of pixels multiplied by the number of clusters. The algorithm is composed of two stages: in the first stage we use the maximum self-organization algorithm until the state of the system converges to the local-minimum, then, the Kohonen self-organization algorithm is used in the last stage in order to improve the solution quality by escaping from the local minimum of the first stage. We have simulated a LANDSAT-TM image data with 500 pixel × 100 pixel image and 8-bit gray scaled. The results justifies all our claims to the proposed algorithm.

KW - Classification

KW - Neural network

KW - Remote-sensing

KW - Self-organization

UR - http://www.scopus.com/inward/record.url?scp=0031681568&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031681568&partnerID=8YFLogxK

M3 - Article

VL - E81-D

SP - 132

EP - 136

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 1

ER -