Data mining in remotely sensed images

a general model and an application

Leen-Kiat Soh, Costas Tsatsoulis

Research output: Contribution to conferencePaper

4 Citations (Scopus)

Abstract

In this paper, we describe a general data mining model for investigating remotely sensed imagery data. We use data mining methodologies to automatically identify patterns (or segmentation classes) and their respective attributes within the image, thus enabling a complete and meaningful description of all pixels. The Data Investigation Model for Unsupervised Segmentation (DIMUS) consists of five modules: inspector, clues generator, classifier, justifier, and mapper. Within the framework of this model, various data mining issues, such as data transformation and selection, information description, determination of segmentation classes, knowledge verification and presentation, are addressed. We have applied the model and implemented a fully automated technique that mines remotely sensed images to learn significant classes or patterns through unsupervised clustering. We have tested successfully the technique on a variety of imagery domains.

Original languageEnglish (US)
Pages798-800
Number of pages3
StatePublished - Jan 1 1998
EventProceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5) - Seattle, WA, USA
Duration: Jul 6 1998Jul 10 1998

Other

OtherProceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5)
CitySeattle, WA, USA
Period7/6/987/10/98

Fingerprint

data mining
Data mining
segmentation
imagery
pixel
Classifiers
Pixels
methodology

ASJC Scopus subject areas

  • Software
  • Geology

Cite this

Soh, L-K., & Tsatsoulis, C. (1998). Data mining in remotely sensed images: a general model and an application. 798-800. Paper presented at Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5), Seattle, WA, USA, .

Data mining in remotely sensed images : a general model and an application. / Soh, Leen-Kiat; Tsatsoulis, Costas.

1998. 798-800 Paper presented at Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5), Seattle, WA, USA, .

Research output: Contribution to conferencePaper

Soh, L-K & Tsatsoulis, C 1998, 'Data mining in remotely sensed images: a general model and an application' Paper presented at Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5), Seattle, WA, USA, 7/6/98 - 7/10/98, pp. 798-800.
Soh L-K, Tsatsoulis C. Data mining in remotely sensed images: a general model and an application. 1998. Paper presented at Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5), Seattle, WA, USA, .
Soh, Leen-Kiat ; Tsatsoulis, Costas. / Data mining in remotely sensed images : a general model and an application. Paper presented at Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5), Seattle, WA, USA, .3 p.
@conference{7e571dfd8f4b4bf5b67029d16c557e68,
title = "Data mining in remotely sensed images: a general model and an application",
abstract = "In this paper, we describe a general data mining model for investigating remotely sensed imagery data. We use data mining methodologies to automatically identify patterns (or segmentation classes) and their respective attributes within the image, thus enabling a complete and meaningful description of all pixels. The Data Investigation Model for Unsupervised Segmentation (DIMUS) consists of five modules: inspector, clues generator, classifier, justifier, and mapper. Within the framework of this model, various data mining issues, such as data transformation and selection, information description, determination of segmentation classes, knowledge verification and presentation, are addressed. We have applied the model and implemented a fully automated technique that mines remotely sensed images to learn significant classes or patterns through unsupervised clustering. We have tested successfully the technique on a variety of imagery domains.",
author = "Leen-Kiat Soh and Costas Tsatsoulis",
year = "1998",
month = "1",
day = "1",
language = "English (US)",
pages = "798--800",
note = "Proceedings of the 1998 IEEE International Geoscience and Remote Sensing Symposium, IGARSS. Part 1 (of 5) ; Conference date: 06-07-1998 Through 10-07-1998",

}

TY - CONF

T1 - Data mining in remotely sensed images

T2 - a general model and an application

AU - Soh, Leen-Kiat

AU - Tsatsoulis, Costas

PY - 1998/1/1

Y1 - 1998/1/1

N2 - In this paper, we describe a general data mining model for investigating remotely sensed imagery data. We use data mining methodologies to automatically identify patterns (or segmentation classes) and their respective attributes within the image, thus enabling a complete and meaningful description of all pixels. The Data Investigation Model for Unsupervised Segmentation (DIMUS) consists of five modules: inspector, clues generator, classifier, justifier, and mapper. Within the framework of this model, various data mining issues, such as data transformation and selection, information description, determination of segmentation classes, knowledge verification and presentation, are addressed. We have applied the model and implemented a fully automated technique that mines remotely sensed images to learn significant classes or patterns through unsupervised clustering. We have tested successfully the technique on a variety of imagery domains.

AB - In this paper, we describe a general data mining model for investigating remotely sensed imagery data. We use data mining methodologies to automatically identify patterns (or segmentation classes) and their respective attributes within the image, thus enabling a complete and meaningful description of all pixels. The Data Investigation Model for Unsupervised Segmentation (DIMUS) consists of five modules: inspector, clues generator, classifier, justifier, and mapper. Within the framework of this model, various data mining issues, such as data transformation and selection, information description, determination of segmentation classes, knowledge verification and presentation, are addressed. We have applied the model and implemented a fully automated technique that mines remotely sensed images to learn significant classes or patterns through unsupervised clustering. We have tested successfully the technique on a variety of imagery domains.

UR - http://www.scopus.com/inward/record.url?scp=0031626217&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031626217&partnerID=8YFLogxK

M3 - Paper

SP - 798

EP - 800

ER -