Using suitable neighbors to augment the training set in hyperspectral maximum likelihood classification

John A. Richards; Xiuping Jia

doi:10.1109/LGRS.2008.2005512

Using suitable neighbors to augment the training set in hyperspectral maximum likelihood classification

John A. Richards, Xiuping Jia

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

55 Citations (Scopus)

Abstract

A method is presented for supplementing the training set in maximum likelihood classification of hyperspectral data to mitigate the Hughes phenomenon. Based on the idea that the near neighbors of training pixels are likely to come from the same class, measures are proposed to assess neighbors as potential candidates so that those selected give improved class statistics and classification accuracy.

Original language	English
Article number	4656467
Pages (from-to)	774-777
Number of pages	4
Journal	IEEE Geoscience and Remote Sensing Letters
Volume	5
Issue number	4
DOIs	https://doi.org/10.1109/LGRS.2008.2005512
Publication status	Published - Oct 2008

Access to Document

10.1109/LGRS.2008.2005512

Cite this

@article{512ee6403fd34b46b7a005fdbb5bbadc,

title = "Using suitable neighbors to augment the training set in hyperspectral maximum likelihood classification",

abstract = "A method is presented for supplementing the training set in maximum likelihood classification of hyperspectral data to mitigate the Hughes phenomenon. Based on the idea that the near neighbors of training pixels are likely to come from the same class, measures are proposed to assess neighbors as potential candidates so that those selected give improved class statistics and classification accuracy.",

keywords = "Classification, Generalization, Hyperspectral, Training",

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

year = "2008",

month = oct,

doi = "10.1109/LGRS.2008.2005512",

language = "English",

volume = "5",

pages = "774--777",

journal = "IEEE Geoscience and Remote Sensing Letters",

issn = "1545-598X",

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

number = "4",

}

TY - JOUR

T1 - Using suitable neighbors to augment the training set in hyperspectral maximum likelihood classification

AU - Richards, John A.

AU - Jia, Xiuping

PY - 2008/10

Y1 - 2008/10

N2 - A method is presented for supplementing the training set in maximum likelihood classification of hyperspectral data to mitigate the Hughes phenomenon. Based on the idea that the near neighbors of training pixels are likely to come from the same class, measures are proposed to assess neighbors as potential candidates so that those selected give improved class statistics and classification accuracy.

AB - A method is presented for supplementing the training set in maximum likelihood classification of hyperspectral data to mitigate the Hughes phenomenon. Based on the idea that the near neighbors of training pixels are likely to come from the same class, measures are proposed to assess neighbors as potential candidates so that those selected give improved class statistics and classification accuracy.

KW - Classification

KW - Generalization

KW - Hyperspectral

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=55649083745&partnerID=8YFLogxK

U2 - 10.1109/LGRS.2008.2005512

DO - 10.1109/LGRS.2008.2005512

M3 - Article

SN - 1545-598X

VL - 5

SP - 774

EP - 777

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

IS - 4

M1 - 4656467

ER -

Using suitable neighbors to augment the training set in hyperspectral maximum likelihood classification

Abstract

Access to Document

Other files and links

Fingerprint

Cite this