Contextual unsupervised classification of remotely sensed imagery with mixels

Shuji Kawaguchi, Ryuei Nishii

研究成果: 著書/レポートタイプへの貢献会議での発言

抄録

We propose a contextual unsupervised classification method of geostatistical data based on combination of Ward clustering method and Markov random fields (MRF). Image is clustered into classes by using not only spectrum of pixels but also spatial information. For the classification of remote sensing data of low spatial resolution, the treatment of mixed pixel is importance. From the knowledge that the most of mixed pixels locate in boundaries of land-covers, we first detect edge pixels and remove them from the image. We here introduce a new measure of spatial adjacency of the classes. Spatial adjacency is used to MRF-based update of the classes. Clustering of edge pixels are processed as final step. It is shown that the proposed method gives higher accuracy than conventional clustering method does.

元の言語英語
ホスト出版物のタイトルImage and Signal Processing for Remote Sensing XII
DOI
出版物ステータス出版済み - 12 1 2006
イベントImage and Signal Processing for Remote Sensing XII - Stockholm, スウェーデン
継続期間: 9 11 20069 14 2006

出版物シリーズ

名前Proceedings of SPIE - The International Society for Optical Engineering
6365
ISSN(印刷物)0277-786X

その他

その他Image and Signal Processing for Remote Sensing XII
スウェーデン
Stockholm
期間9/11/069/14/06

Fingerprint

Unsupervised Classification
imagery
Pixel
Pixels
pixels
Adjacency
Clustering Methods
Random Field
Land Cover
Spatial Information
Spatial Resolution
Remote Sensing
remote sensing
Remote sensing
High Accuracy
spatial resolution
Update
Imagery
Clustering
Class

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

これを引用

Kawaguchi, S., & Nishii, R. (2006). Contextual unsupervised classification of remotely sensed imagery with mixels. : Image and Signal Processing for Remote Sensing XII [63650R] (Proceedings of SPIE - The International Society for Optical Engineering; 巻数 6365). https://doi.org/10.1117/12.689566

Contextual unsupervised classification of remotely sensed imagery with mixels. / Kawaguchi, Shuji; Nishii, Ryuei.

Image and Signal Processing for Remote Sensing XII. 2006. 63650R (Proceedings of SPIE - The International Society for Optical Engineering; 巻 6365).

研究成果: 著書/レポートタイプへの貢献会議での発言

Kawaguchi, S & Nishii, R 2006, Contextual unsupervised classification of remotely sensed imagery with mixels. : Image and Signal Processing for Remote Sensing XII., 63650R, Proceedings of SPIE - The International Society for Optical Engineering, 巻. 6365, Image and Signal Processing for Remote Sensing XII, Stockholm, スウェーデン, 9/11/06. https://doi.org/10.1117/12.689566
Kawaguchi S, Nishii R. Contextual unsupervised classification of remotely sensed imagery with mixels. : Image and Signal Processing for Remote Sensing XII. 2006. 63650R. (Proceedings of SPIE - The International Society for Optical Engineering). https://doi.org/10.1117/12.689566
Kawaguchi, Shuji ; Nishii, Ryuei. / Contextual unsupervised classification of remotely sensed imagery with mixels. Image and Signal Processing for Remote Sensing XII. 2006. (Proceedings of SPIE - The International Society for Optical Engineering).
@inproceedings{0db4b285bad448bb872d4cafca851b0e,
title = "Contextual unsupervised classification of remotely sensed imagery with mixels",
abstract = "We propose a contextual unsupervised classification method of geostatistical data based on combination of Ward clustering method and Markov random fields (MRF). Image is clustered into classes by using not only spectrum of pixels but also spatial information. For the classification of remote sensing data of low spatial resolution, the treatment of mixed pixel is importance. From the knowledge that the most of mixed pixels locate in boundaries of land-covers, we first detect edge pixels and remove them from the image. We here introduce a new measure of spatial adjacency of the classes. Spatial adjacency is used to MRF-based update of the classes. Clustering of edge pixels are processed as final step. It is shown that the proposed method gives higher accuracy than conventional clustering method does.",
author = "Shuji Kawaguchi and Ryuei Nishii",
year = "2006",
month = "12",
day = "1",
doi = "10.1117/12.689566",
language = "English",
isbn = "0819464600",
series = "Proceedings of SPIE - The International Society for Optical Engineering",
booktitle = "Image and Signal Processing for Remote Sensing XII",

}

TY - GEN

T1 - Contextual unsupervised classification of remotely sensed imagery with mixels

AU - Kawaguchi, Shuji

AU - Nishii, Ryuei

PY - 2006/12/1

Y1 - 2006/12/1

N2 - We propose a contextual unsupervised classification method of geostatistical data based on combination of Ward clustering method and Markov random fields (MRF). Image is clustered into classes by using not only spectrum of pixels but also spatial information. For the classification of remote sensing data of low spatial resolution, the treatment of mixed pixel is importance. From the knowledge that the most of mixed pixels locate in boundaries of land-covers, we first detect edge pixels and remove them from the image. We here introduce a new measure of spatial adjacency of the classes. Spatial adjacency is used to MRF-based update of the classes. Clustering of edge pixels are processed as final step. It is shown that the proposed method gives higher accuracy than conventional clustering method does.

AB - We propose a contextual unsupervised classification method of geostatistical data based on combination of Ward clustering method and Markov random fields (MRF). Image is clustered into classes by using not only spectrum of pixels but also spatial information. For the classification of remote sensing data of low spatial resolution, the treatment of mixed pixel is importance. From the knowledge that the most of mixed pixels locate in boundaries of land-covers, we first detect edge pixels and remove them from the image. We here introduce a new measure of spatial adjacency of the classes. Spatial adjacency is used to MRF-based update of the classes. Clustering of edge pixels are processed as final step. It is shown that the proposed method gives higher accuracy than conventional clustering method does.

UR - http://www.scopus.com/inward/record.url?scp=33751424141&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33751424141&partnerID=8YFLogxK

U2 - 10.1117/12.689566

DO - 10.1117/12.689566

M3 - Conference contribution

AN - SCOPUS:33751424141

SN - 0819464600

SN - 9780819464606

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Image and Signal Processing for Remote Sensing XII

ER -