Clustering of the Poincare vectors

Author

Lakroum, S. ; Devlaminck, V. ; Terrier, P. ; Biela-Enberg, P. ; Postaire, J.-G.

Author_Institution

Lab. d´´Automatique, de Genie Inf. et Signal, Univ. des Sci. et Technol. de Lille, France

Volume

2

fYear

2005

fDate

11-14 Sept. 2005

Abstract

Data clustering is useful for discovering significant patterns and characteristics in large datasets. In this paper, we address the problem of clustering the Poincare vectors. Three variants of the k-means algorithm and a competitive neural technique are tested and compared. The empirical performance of the different methods in terms of classification and segmentation accuracy is evaluated. The results obtained on real life passive polarimetric images demonstrate the usefulness of such approach for exploiting the polarimetric information.

Keywords

image classification; image segmentation; matrix algebra; pattern clustering; polarimetry; unsupervised learning; vectors; Poincare vectors; classification accuracy; competitive neural technique; data clustering; k-means algorithm; passive polarimetric images; segmentation accuracy; unsupervised classification; Clustering algorithms; Electromagnetic wave polarization; Image segmentation; Layout; Optical imaging; Optical polarization; Optical surface waves; Polarimetry; Stokes parameters; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 2005. ICIP 2005. IEEE International Conference on

Print_ISBN

0-7803-9134-9

Type

conf

DOI

10.1109/ICIP.2005.1530274

Filename

1530274