Separability measures for error estimation of two normally distributed classes

Joonyong Hong, Chulhee Lee

Research output: Contribution to journalConference article

1 Citation (Scopus)

Abstract

In pattern classification and remote sensing, the Gaussian ML classifier is most widely used because of its speed and robustness. In this paper, we propose to use two separability measures, Bhattacharyya distance and divergence to estimate the classification error of the Gaussian ML classifier. In the proposed method, we try to find empirical relationship between the separability measures and the classification error. In order to find such relationship, we generate two classes with normal distribution and compute the separability measures and classification error between the classes. Although there are infinite number of possibilities that two classes can have, we systematically search the whole mean-covariance space. From this exhaustive search, we are able to estimate the classification error accurately using the Bhattacharyya distance and divergence. It is observed that the error estimation using both the Bhattacharyya distance and divergence does not give a significant improvement over the error estimation using the Bhattacharyya distance only.

Original languageEnglish
Pages (from-to)695-705
Number of pages11
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume3460
DOIs
Publication statusPublished - 1998 Dec 1

Fingerprint

Error Estimation
Separability
Error analysis
Divergence
divergence
Classifiers
Classifier
classifiers
Divergence Measure
Pattern Classification
Exhaustive Search
Distance Measure
Normal distribution
Remote Sensing
Estimate
Pattern recognition
Gaussian distribution
Remote sensing
estimates
Robustness

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

@article{a7e8c79551274a8a8fe0249fa29f91dc,
title = "Separability measures for error estimation of two normally distributed classes",
abstract = "In pattern classification and remote sensing, the Gaussian ML classifier is most widely used because of its speed and robustness. In this paper, we propose to use two separability measures, Bhattacharyya distance and divergence to estimate the classification error of the Gaussian ML classifier. In the proposed method, we try to find empirical relationship between the separability measures and the classification error. In order to find such relationship, we generate two classes with normal distribution and compute the separability measures and classification error between the classes. Although there are infinite number of possibilities that two classes can have, we systematically search the whole mean-covariance space. From this exhaustive search, we are able to estimate the classification error accurately using the Bhattacharyya distance and divergence. It is observed that the error estimation using both the Bhattacharyya distance and divergence does not give a significant improvement over the error estimation using the Bhattacharyya distance only.",
author = "Joonyong Hong and Chulhee Lee",
year = "1998",
month = "12",
day = "1",
doi = "10.1117/12.323225",
language = "English",
volume = "3460",
pages = "695--705",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

Separability measures for error estimation of two normally distributed classes. / Hong, Joonyong; Lee, Chulhee.

In: Proceedings of SPIE - The International Society for Optical Engineering, Vol. 3460, 01.12.1998, p. 695-705.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Separability measures for error estimation of two normally distributed classes

AU - Hong, Joonyong

AU - Lee, Chulhee

PY - 1998/12/1

Y1 - 1998/12/1

N2 - In pattern classification and remote sensing, the Gaussian ML classifier is most widely used because of its speed and robustness. In this paper, we propose to use two separability measures, Bhattacharyya distance and divergence to estimate the classification error of the Gaussian ML classifier. In the proposed method, we try to find empirical relationship between the separability measures and the classification error. In order to find such relationship, we generate two classes with normal distribution and compute the separability measures and classification error between the classes. Although there are infinite number of possibilities that two classes can have, we systematically search the whole mean-covariance space. From this exhaustive search, we are able to estimate the classification error accurately using the Bhattacharyya distance and divergence. It is observed that the error estimation using both the Bhattacharyya distance and divergence does not give a significant improvement over the error estimation using the Bhattacharyya distance only.

AB - In pattern classification and remote sensing, the Gaussian ML classifier is most widely used because of its speed and robustness. In this paper, we propose to use two separability measures, Bhattacharyya distance and divergence to estimate the classification error of the Gaussian ML classifier. In the proposed method, we try to find empirical relationship between the separability measures and the classification error. In order to find such relationship, we generate two classes with normal distribution and compute the separability measures and classification error between the classes. Although there are infinite number of possibilities that two classes can have, we systematically search the whole mean-covariance space. From this exhaustive search, we are able to estimate the classification error accurately using the Bhattacharyya distance and divergence. It is observed that the error estimation using both the Bhattacharyya distance and divergence does not give a significant improvement over the error estimation using the Bhattacharyya distance only.

UR - http://www.scopus.com/inward/record.url?scp=0032402416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032402416&partnerID=8YFLogxK

U2 - 10.1117/12.323225

DO - 10.1117/12.323225

M3 - Conference article

AN - SCOPUS:0032402416

VL - 3460

SP - 695

EP - 705

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

ER -