Incorporating Knowledge into Unsupervised Model-Based Clustering for Satellite Image

B Al_Momani, PJ Morrow, SI McClean

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The identification and classification of landcover types from remotely sensed data is traditionally based on the assumption that pixels with similar spatial distribution patterns belong to the same spectral class. However, spectral data on its own has proven to be insufficient for classification. In addition, it is difficult to obtain enough accurate labelled samples from such data. Contextual data can be incorporated or fused' with spectral data to improve the estimation of class labels and therefore enhance the accuracy of the classification process as a whole when labelled data is not available. In this paper we use Dempster-Shafer theory of evidence to fuse the output of an unsupervised model-based clustering (MBC) technique and contextual data in the form of a digital elevation model. The final classification accuracy is shown to improve when using this approach.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages746-753
Number of pages8
DOIs
Publication statusPublished - 13 May 2007
EventIEEE/ASC International Conference on Computer Systems and Applications, AICCSA07 - Amman-Jordan
Duration: 13 May 2007 → …

Conference

ConferenceIEEE/ASC International Conference on Computer Systems and Applications, AICCSA07
Period13/05/07 → …

Fingerprint

satellite image
digital elevation model
pixel
land cover
spatial distribution

Cite this

@inproceedings{9433417625e74731acef2139a3380799,
title = "Incorporating Knowledge into Unsupervised Model-Based Clustering for Satellite Image",
abstract = "The identification and classification of landcover types from remotely sensed data is traditionally based on the assumption that pixels with similar spatial distribution patterns belong to the same spectral class. However, spectral data on its own has proven to be insufficient for classification. In addition, it is difficult to obtain enough accurate labelled samples from such data. Contextual data can be incorporated or fused' with spectral data to improve the estimation of class labels and therefore enhance the accuracy of the classification process as a whole when labelled data is not available. In this paper we use Dempster-Shafer theory of evidence to fuse the output of an unsupervised model-based clustering (MBC) technique and contextual data in the form of a digital elevation model. The final classification accuracy is shown to improve when using this approach.",
author = "B Al_Momani and PJ Morrow and SI McClean",
year = "2007",
month = "5",
day = "13",
doi = "10.1109/AICCSA.2007.370716",
language = "English",
isbn = "1-4244-1030-4",
pages = "746--753",
booktitle = "Unknown Host Publication",

}

Al_Momani, B, Morrow, PJ & McClean, SI 2007, Incorporating Knowledge into Unsupervised Model-Based Clustering for Satellite Image. in Unknown Host Publication. pp. 746-753, IEEE/ASC International Conference on Computer Systems and Applications, AICCSA07, 13/05/07. https://doi.org/10.1109/AICCSA.2007.370716

Incorporating Knowledge into Unsupervised Model-Based Clustering for Satellite Image. / Al_Momani, B; Morrow, PJ; McClean, SI.

Unknown Host Publication. 2007. p. 746-753.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Incorporating Knowledge into Unsupervised Model-Based Clustering for Satellite Image

AU - Al_Momani, B

AU - Morrow, PJ

AU - McClean, SI

PY - 2007/5/13

Y1 - 2007/5/13

N2 - The identification and classification of landcover types from remotely sensed data is traditionally based on the assumption that pixels with similar spatial distribution patterns belong to the same spectral class. However, spectral data on its own has proven to be insufficient for classification. In addition, it is difficult to obtain enough accurate labelled samples from such data. Contextual data can be incorporated or fused' with spectral data to improve the estimation of class labels and therefore enhance the accuracy of the classification process as a whole when labelled data is not available. In this paper we use Dempster-Shafer theory of evidence to fuse the output of an unsupervised model-based clustering (MBC) technique and contextual data in the form of a digital elevation model. The final classification accuracy is shown to improve when using this approach.

AB - The identification and classification of landcover types from remotely sensed data is traditionally based on the assumption that pixels with similar spatial distribution patterns belong to the same spectral class. However, spectral data on its own has proven to be insufficient for classification. In addition, it is difficult to obtain enough accurate labelled samples from such data. Contextual data can be incorporated or fused' with spectral data to improve the estimation of class labels and therefore enhance the accuracy of the classification process as a whole when labelled data is not available. In this paper we use Dempster-Shafer theory of evidence to fuse the output of an unsupervised model-based clustering (MBC) technique and contextual data in the form of a digital elevation model. The final classification accuracy is shown to improve when using this approach.

U2 - 10.1109/AICCSA.2007.370716

DO - 10.1109/AICCSA.2007.370716

M3 - Conference contribution

SN - 1-4244-1030-4

SP - 746

EP - 753

BT - Unknown Host Publication

ER -