Fast k-NN classification using the cluster-space approach

Xiuping Jia; John A. Richards

doi:10.1109/LGRS.2005.846437

Fast k-NN classification using the cluster-space approach

Xiuping Jia^*, John A. Richards

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

33 Citations (Scopus)

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.

Original language	English
Pages (from-to)	225-228
Number of pages	4
Journal	IEEE Geoscience and Remote Sensing Letters
Volume	2
Issue number	2
DOIs	https://doi.org/10.1109/LGRS.2005.846437
Publication status	Published - Apr 2005

Access to Document

10.1109/LGRS.2005.846437

Cite this

@article{3fa4d7567d9743e88f9b19ff1f3d3e4e,

title = "Fast k-NN classification using the cluster-space approach",

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 = "Classification, Cluster-space representation, Hyperspectral, K-NN",

author = "Xiuping Jia and Richards, {John A.}",

year = "2005",

month = apr,

doi = "10.1109/LGRS.2005.846437",

language = "English",

volume = "2",

pages = "225--228",

journal = "IEEE Geoscience and Remote Sensing Letters",

issn = "1545-598X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Fast k-NN classification using the cluster-space approach

AU - Jia, Xiuping

AU - Richards, John A.

PY - 2005/4

Y1 - 2005/4

N2 - 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.

AB - 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.

KW - Classification

KW - Cluster-space representation

KW - Hyperspectral

KW - K-NN

UR - http://www.scopus.com/inward/record.url?scp=18844447262&partnerID=8YFLogxK

U2 - 10.1109/LGRS.2005.846437

DO - 10.1109/LGRS.2005.846437

M3 - Article

SN - 1545-598X

VL - 2

SP - 225

EP - 228

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

IS - 2

ER -

Fast k-NN classification using the cluster-space approach

Abstract

Access to Document

Other files and links

Fingerprint

Cite this