Building classification of terrestrial images by generic geometric hierarchical cluster analysis features

Gerd Brunner, Hans Burkhardt

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

Abstract

The scope of this paper is the challenging task of classifying terrestrial images of buildings, automatically. Straight line segments and their connectivity incorporate significant information about object shapes. Man-made buildings exhibit special generic shapes which are extracted from embedded spatial and angular line segment relationships by cluster analysis. After employing an agglomerative hierarchical cluster analysis we obtain geometrical structure information features on different scales. For the classification process we apply support vector machines (SVM) with polynomial and radial basis function (RBF) kernels to separate the feature space by a hyperplane into 2 classes. The method is applied to an image collection taken from the Corel image database and compared with traditional edge- orientation histogram features. We obtained a 88 % true positive classification rate (recall) with an F-measure value of 81.3 %.

Original languageEnglish (US)
Title of host publicationProceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005
Pages136-139
Number of pages4
StatePublished - Dec 1 2005
Event9th IAPR Conference on Machine Vision Applications, MVA 2005 - Tsukuba Science City, Japan
Duration: May 16 2005May 18 2005

Publication series

NameProceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005

Conference

Conference9th IAPR Conference on Machine Vision Applications, MVA 2005
CountryJapan
CityTsukuba Science City
Period5/16/055/18/05

Fingerprint

Cluster analysis
Support vector machines
Polynomials

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Brunner, G., & Burkhardt, H. (2005). Building classification of terrestrial images by generic geometric hierarchical cluster analysis features. In Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005 (pp. 136-139). (Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005).
Brunner, Gerd ; Burkhardt, Hans. / Building classification of terrestrial images by generic geometric hierarchical cluster analysis features. Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005. 2005. pp. 136-139 (Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005).
@inproceedings{7de4445497bb4659a21243651aa14d77,
title = "Building classification of terrestrial images by generic geometric hierarchical cluster analysis features",
abstract = "The scope of this paper is the challenging task of classifying terrestrial images of buildings, automatically. Straight line segments and their connectivity incorporate significant information about object shapes. Man-made buildings exhibit special generic shapes which are extracted from embedded spatial and angular line segment relationships by cluster analysis. After employing an agglomerative hierarchical cluster analysis we obtain geometrical structure information features on different scales. For the classification process we apply support vector machines (SVM) with polynomial and radial basis function (RBF) kernels to separate the feature space by a hyperplane into 2 classes. The method is applied to an image collection taken from the Corel image database and compared with traditional edge- orientation histogram features. We obtained a 88 {\%} true positive classification rate (recall) with an F-measure value of 81.3 {\%}.",
author = "Gerd Brunner and Hans Burkhardt",
year = "2005",
month = "12",
day = "1",
language = "English (US)",
isbn = "4901122045",
series = "Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005",
pages = "136--139",
booktitle = "Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005",

}

Brunner, G & Burkhardt, H 2005, Building classification of terrestrial images by generic geometric hierarchical cluster analysis features. in Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005. Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005, pp. 136-139, 9th IAPR Conference on Machine Vision Applications, MVA 2005, Tsukuba Science City, Japan, 5/16/05.

Building classification of terrestrial images by generic geometric hierarchical cluster analysis features. / Brunner, Gerd; Burkhardt, Hans.

Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005. 2005. p. 136-139 (Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005).

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

TY - GEN

T1 - Building classification of terrestrial images by generic geometric hierarchical cluster analysis features

AU - Brunner, Gerd

AU - Burkhardt, Hans

PY - 2005/12/1

Y1 - 2005/12/1

N2 - The scope of this paper is the challenging task of classifying terrestrial images of buildings, automatically. Straight line segments and their connectivity incorporate significant information about object shapes. Man-made buildings exhibit special generic shapes which are extracted from embedded spatial and angular line segment relationships by cluster analysis. After employing an agglomerative hierarchical cluster analysis we obtain geometrical structure information features on different scales. For the classification process we apply support vector machines (SVM) with polynomial and radial basis function (RBF) kernels to separate the feature space by a hyperplane into 2 classes. The method is applied to an image collection taken from the Corel image database and compared with traditional edge- orientation histogram features. We obtained a 88 % true positive classification rate (recall) with an F-measure value of 81.3 %.

AB - The scope of this paper is the challenging task of classifying terrestrial images of buildings, automatically. Straight line segments and their connectivity incorporate significant information about object shapes. Man-made buildings exhibit special generic shapes which are extracted from embedded spatial and angular line segment relationships by cluster analysis. After employing an agglomerative hierarchical cluster analysis we obtain geometrical structure information features on different scales. For the classification process we apply support vector machines (SVM) with polynomial and radial basis function (RBF) kernels to separate the feature space by a hyperplane into 2 classes. The method is applied to an image collection taken from the Corel image database and compared with traditional edge- orientation histogram features. We obtained a 88 % true positive classification rate (recall) with an F-measure value of 81.3 %.

UR - http://www.scopus.com/inward/record.url?scp=84872579123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872579123&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84872579123

SN - 4901122045

SN - 9784901122047

T3 - Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005

SP - 136

EP - 139

BT - Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005

ER -

Brunner G, Burkhardt H. Building classification of terrestrial images by generic geometric hierarchical cluster analysis features. In Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005. 2005. p. 136-139. (Proceedings of the 9th IAPR Conference on Machine Vision Applications, MVA 2005).