Target classification using Gaussian mixture model for ground surveillance Doppler radar

Igal Bilik; Joseph Tabrikian; Arnon Cohen

doi:10.1109/RADAR.2005.1435957

Target classification using Gaussian mixture model for ground surveillance Doppler radar

Igal Bilik, Joseph Tabrikian, Arnon Cohen

Department of Electrical & Computer Engineering

Research output: Contribution to journal › Conference article › peer-review

10 Scopus citations

Abstract

An automatic target recognition (ATR) algorithm, based on greedy learning of Gaussian mixture model (GMM) is developed in this work. The GMMs were obtained for a wide range of ground surveillance radar targets such as: walking person(s), tracked or wheeled vehicles, animals and clutter. Maximum-likelihood (ML) and "majority voting" decision schemes were applied to these models for target classification. The corresponding classifiers were trained and tested using distinct databases of target echoes, recorded by ground surveillance radar. ML and "majority voting" classifiers obtained classification rates of 88% and 96%, correspondingly. Both classifiers outperform trained human operators.

Original language	English
Article number	1435957
Pages (from-to)	910-915
Number of pages	6
Journal	IEEE National Radar Conference - Proceedings
Volume	2005-January
Issue number	January
DOIs	https://doi.org/10.1109/RADAR.2005.1435957
State	Published - 1 Jan 2005
Event	2005 IEEE International Radar Conference Record, RADAR 2005 - Arlington, United States Duration: 9 May 2005 → 12 May 2005

Keywords

ATR
GMM
Ground surveillance radar
Majority voting
Target classification

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1109/RADAR.2005.1435957

Cite this

@article{4430f466e81b471e937d7105c0975ef8,

title = "Target classification using Gaussian mixture model for ground surveillance Doppler radar",

abstract = "An automatic target recognition (ATR) algorithm, based on greedy learning of Gaussian mixture model (GMM) is developed in this work. The GMMs were obtained for a wide range of ground surveillance radar targets such as: walking person(s), tracked or wheeled vehicles, animals and clutter. Maximum-likelihood (ML) and {"}majority voting{"} decision schemes were applied to these models for target classification. The corresponding classifiers were trained and tested using distinct databases of target echoes, recorded by ground surveillance radar. ML and {"}majority voting{"} classifiers obtained classification rates of 88% and 96%, correspondingly. Both classifiers outperform trained human operators.",

keywords = "ATR, GMM, Ground surveillance radar, Majority voting, Target classification",

author = "Igal Bilik and Joseph Tabrikian and Arnon Cohen",

note = "Publisher Copyright: {\textcopyright} 2005 IEEE.; 2005 IEEE International Radar Conference Record, RADAR 2005 ; Conference date: 09-05-2005 Through 12-05-2005",

year = "2005",

month = jan,

day = "1",

doi = "10.1109/RADAR.2005.1435957",

language = "English",

volume = "2005-January",

pages = "910--915",

journal = "IEEE National Radar Conference - Proceedings",

issn = "1097-5659",

publisher = "Institute of Electrical and Electronics Engineers",

number = "January",

}

TY - JOUR

T1 - Target classification using Gaussian mixture model for ground surveillance Doppler radar

AU - Bilik, Igal

AU - Tabrikian, Joseph

AU - Cohen, Arnon

PY - 2005/1/1

Y1 - 2005/1/1

N2 - An automatic target recognition (ATR) algorithm, based on greedy learning of Gaussian mixture model (GMM) is developed in this work. The GMMs were obtained for a wide range of ground surveillance radar targets such as: walking person(s), tracked or wheeled vehicles, animals and clutter. Maximum-likelihood (ML) and "majority voting" decision schemes were applied to these models for target classification. The corresponding classifiers were trained and tested using distinct databases of target echoes, recorded by ground surveillance radar. ML and "majority voting" classifiers obtained classification rates of 88% and 96%, correspondingly. Both classifiers outperform trained human operators.

AB - An automatic target recognition (ATR) algorithm, based on greedy learning of Gaussian mixture model (GMM) is developed in this work. The GMMs were obtained for a wide range of ground surveillance radar targets such as: walking person(s), tracked or wheeled vehicles, animals and clutter. Maximum-likelihood (ML) and "majority voting" decision schemes were applied to these models for target classification. The corresponding classifiers were trained and tested using distinct databases of target echoes, recorded by ground surveillance radar. ML and "majority voting" classifiers obtained classification rates of 88% and 96%, correspondingly. Both classifiers outperform trained human operators.

KW - ATR

KW - GMM

KW - Ground surveillance radar

KW - Majority voting

KW - Target classification

UR - http://www.scopus.com/inward/record.url?scp=57649164828&partnerID=8YFLogxK

U2 - 10.1109/RADAR.2005.1435957

DO - 10.1109/RADAR.2005.1435957

M3 - Conference article

AN - SCOPUS:57649164828

SN - 1097-5659

VL - 2005-January

SP - 910

EP - 915

JO - IEEE National Radar Conference - Proceedings

JF - IEEE National Radar Conference - Proceedings

IS - January

M1 - 1435957

T2 - 2005 IEEE International Radar Conference Record, RADAR 2005

Y2 - 9 May 2005 through 12 May 2005

ER -

Target classification using Gaussian mixture model for ground surveillance Doppler radar

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this