Feature extraction based on difference vectors

Taeuk Jeong, Jong Geun Park, Chulhee Lee

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

1 Citation (Scopus)

Abstract

In a typical classification procedure of high dimensional data, feature extraction is first applied to reduce the dimensionality and a classifier is employed. However, in most feature extraction methods, covariance matrices must be estimated. When training samples are limited, this estimation is inherently biased, thereby generating ineffective features. In this paper, we propose a new feature extraction method for high dimensional hyperspectral data when limited training samples are available. In the proposed method, we construct a feature matrix using available training samples. The proposed method calculates the difference vector feature matrix using weighted difference vectors among the training samples. Experimental results show that the proposed method improves classification accuracy even if the size of training sample is very small.

Original languageEnglish
Title of host publication2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007
Pages183-186
Number of pages4
DOIs
Publication statusPublished - 2007 Dec 1
Event2nd IEEE International Workshop on Soft Computing Applications, SOFA 2007 - Oradea, Romania
Duration: 2007 Aug 212007 Aug 23

Other

Other2nd IEEE International Workshop on Soft Computing Applications, SOFA 2007
CountryRomania
CityOradea
Period07/8/2107/8/23

Fingerprint

Feature extraction
Covariance matrix
Classifiers

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Software

Cite this

Jeong, T., Park, J. G., & Lee, C. (2007). Feature extraction based on difference vectors. In 2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007 (pp. 183-186). [4318325] https://doi.org/10.1109/SOFA.2007.4318325
Jeong, Taeuk ; Park, Jong Geun ; Lee, Chulhee. / Feature extraction based on difference vectors. 2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007. 2007. pp. 183-186
@inproceedings{aecaf0ec470d47bab40f1ad6e784251b,
title = "Feature extraction based on difference vectors",
abstract = "In a typical classification procedure of high dimensional data, feature extraction is first applied to reduce the dimensionality and a classifier is employed. However, in most feature extraction methods, covariance matrices must be estimated. When training samples are limited, this estimation is inherently biased, thereby generating ineffective features. In this paper, we propose a new feature extraction method for high dimensional hyperspectral data when limited training samples are available. In the proposed method, we construct a feature matrix using available training samples. The proposed method calculates the difference vector feature matrix using weighted difference vectors among the training samples. Experimental results show that the proposed method improves classification accuracy even if the size of training sample is very small.",
author = "Taeuk Jeong and Park, {Jong Geun} and Chulhee Lee",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/SOFA.2007.4318325",
language = "English",
isbn = "9781424416080",
pages = "183--186",
booktitle = "2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007",

}

Jeong, T, Park, JG & Lee, C 2007, Feature extraction based on difference vectors. in 2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007., 4318325, pp. 183-186, 2nd IEEE International Workshop on Soft Computing Applications, SOFA 2007, Oradea, Romania, 07/8/21. https://doi.org/10.1109/SOFA.2007.4318325

Feature extraction based on difference vectors. / Jeong, Taeuk; Park, Jong Geun; Lee, Chulhee.

2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007. 2007. p. 183-186 4318325.

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

TY - GEN

T1 - Feature extraction based on difference vectors

AU - Jeong, Taeuk

AU - Park, Jong Geun

AU - Lee, Chulhee

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In a typical classification procedure of high dimensional data, feature extraction is first applied to reduce the dimensionality and a classifier is employed. However, in most feature extraction methods, covariance matrices must be estimated. When training samples are limited, this estimation is inherently biased, thereby generating ineffective features. In this paper, we propose a new feature extraction method for high dimensional hyperspectral data when limited training samples are available. In the proposed method, we construct a feature matrix using available training samples. The proposed method calculates the difference vector feature matrix using weighted difference vectors among the training samples. Experimental results show that the proposed method improves classification accuracy even if the size of training sample is very small.

AB - In a typical classification procedure of high dimensional data, feature extraction is first applied to reduce the dimensionality and a classifier is employed. However, in most feature extraction methods, covariance matrices must be estimated. When training samples are limited, this estimation is inherently biased, thereby generating ineffective features. In this paper, we propose a new feature extraction method for high dimensional hyperspectral data when limited training samples are available. In the proposed method, we construct a feature matrix using available training samples. The proposed method calculates the difference vector feature matrix using weighted difference vectors among the training samples. Experimental results show that the proposed method improves classification accuracy even if the size of training sample is very small.

UR - http://www.scopus.com/inward/record.url?scp=47949113084&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47949113084&partnerID=8YFLogxK

U2 - 10.1109/SOFA.2007.4318325

DO - 10.1109/SOFA.2007.4318325

M3 - Conference contribution

AN - SCOPUS:47949113084

SN - 9781424416080

SP - 183

EP - 186

BT - 2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007

ER -

Jeong T, Park JG, Lee C. Feature extraction based on difference vectors. In 2nd IEEE International Workshop on Soft Computing Applications Proceedings, SOFA 2007. 2007. p. 183-186. 4318325 https://doi.org/10.1109/SOFA.2007.4318325