Fast k-NN classification using the cluster-space approach

Author

Jia, Xiuping ; Richards, John A.

Author_Institution

Sch. of Electr. Eng., Australian Defence Force Acad., ACT, Australia

Volume

2

Issue

2

fYear

2005

fDate

4/1/2005 12:00:00 AM

Firstpage

225

Lastpage

228

Abstract

A fast k-nearest neighbor algorithm is presented which combines k-NN with a cluster-space data representation. Implementation of the algorithm is easier, and classification time can be significantly reduced. Computer-generated data show the modified k-NN retains the advantage of nonparametric analysis but with significant reduction in computational load. Results from tests carried out with Hyperion data demonstrate that the simplification has little effect on classification performance, and yet efficiency is greatly improved.

Keywords

geophysical signal processing; image classification; remote sensing; Hyperion data; cluster-space representation; fast k-NN classification; fast k-nearest neighbor algorithm; geophysical signal processing; hyperspectral imaging; image classification; nonparametric analysis; remote sensing; Bayesian methods; Clustering algorithms; Density functional theory; Hyperspectral imaging; Hyperspectral sensors; Image classification; Nearest neighbor searches; Parameter estimation; Pixel; Testing; Classification; cluster-space representation; hyperspectral;

fLanguage

English

Journal_Title

Geoscience and Remote Sensing Letters, IEEE

Publisher

ieee

ISSN

1545-598X

Type

jour

DOI

10.1109/LGRS.2005.846437

Filename

1420310