SVM based target classification using RCS feature vectors

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

3 Citations (Scopus)

Abstract

This paper investigates the application of SVM (Support Vector Machines) for the classification of stationary human targets and indoor clutter via spectral features. Applying Finite Difference Time Domain (FDTD) techniques allows us to examine the radar cross section (RCS) of humans and indoor clutter objects by utilizing different types of computer models. FDTD allows for the spectral characteristics to be acquired over a wide range of frequencies, polarizations, aspect angles, and materials. The acquired target and clutter RCS spectral characteristics are then investigated in terms of their potential for target classification using SVMs. Based upon variables such as frequency and polarization, a SVM classifier can be trained to classify unknown targets as a human or clutter. Furthermore, the application of feature selection is applied to the spectral characteristics to determine the SVM classification accuracy of a reduced dataset. Classification accuracies of nearly 90% are achieved using radial and polynomial kernels.

Original languageEnglish (US)
Title of host publicationRadar Sensor Technology XIX; and Active and Passive Signatures VI
EditorsArmin Doerry, Chadwick Todd Hawley, G. Charmaine Gilbreath, Kenneth I. Ranney
PublisherSPIE
ISBN (Electronic)9781628415773
DOIs
StatePublished - Jan 1 2015
EventRadar Sensor Technology XIX; and Active and Passive Signatures VI - Baltimore, United States
Duration: Apr 20 2015Apr 23 2015

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9461
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Other

OtherRadar Sensor Technology XIX; and Active and Passive Signatures VI
CountryUnited States
CityBaltimore
Period4/20/154/23/15

Fingerprint

Radar Cross Section
radar cross sections
Radar cross section
clutter
Clutter
Feature Vector
Support vector machines
Support Vector Machine
Target
Finite-difference Time-domain (FDTD)
Polarization
Computer Model
polarization
classifiers
Feature Selection
Feature extraction
polynomials
Classifiers
Classify
Classifier

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Bufler, T. D., Narayanan, R. M., & Dogaru, T. (2015). SVM based target classification using RCS feature vectors. In A. Doerry, C. T. Hawley, G. C. Gilbreath, & K. I. Ranney (Eds.), Radar Sensor Technology XIX; and Active and Passive Signatures VI [94610I] (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 9461). SPIE. https://doi.org/10.1117/12.2176759
Bufler, Travis Dale ; Narayanan, Ram Mohan ; Dogaru, Traian. / SVM based target classification using RCS feature vectors. Radar Sensor Technology XIX; and Active and Passive Signatures VI. editor / Armin Doerry ; Chadwick Todd Hawley ; G. Charmaine Gilbreath ; Kenneth I. Ranney. SPIE, 2015. (Proceedings of SPIE - The International Society for Optical Engineering).
@inproceedings{f8ed4e0c0dac43fb803db0788cd0a983,
title = "SVM based target classification using RCS feature vectors",
abstract = "This paper investigates the application of SVM (Support Vector Machines) for the classification of stationary human targets and indoor clutter via spectral features. Applying Finite Difference Time Domain (FDTD) techniques allows us to examine the radar cross section (RCS) of humans and indoor clutter objects by utilizing different types of computer models. FDTD allows for the spectral characteristics to be acquired over a wide range of frequencies, polarizations, aspect angles, and materials. The acquired target and clutter RCS spectral characteristics are then investigated in terms of their potential for target classification using SVMs. Based upon variables such as frequency and polarization, a SVM classifier can be trained to classify unknown targets as a human or clutter. Furthermore, the application of feature selection is applied to the spectral characteristics to determine the SVM classification accuracy of a reduced dataset. Classification accuracies of nearly 90{\%} are achieved using radial and polynomial kernels.",
author = "Bufler, {Travis Dale} and Narayanan, {Ram Mohan} and Traian Dogaru",
year = "2015",
month = "1",
day = "1",
doi = "10.1117/12.2176759",
language = "English (US)",
series = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",
editor = "Armin Doerry and Hawley, {Chadwick Todd} and Gilbreath, {G. Charmaine} and Ranney, {Kenneth I.}",
booktitle = "Radar Sensor Technology XIX; and Active and Passive Signatures VI",
address = "United States",

}

Bufler, TD, Narayanan, RM & Dogaru, T 2015, SVM based target classification using RCS feature vectors. in A Doerry, CT Hawley, GC Gilbreath & KI Ranney (eds), Radar Sensor Technology XIX; and Active and Passive Signatures VI., 94610I, Proceedings of SPIE - The International Society for Optical Engineering, vol. 9461, SPIE, Radar Sensor Technology XIX; and Active and Passive Signatures VI, Baltimore, United States, 4/20/15. https://doi.org/10.1117/12.2176759

SVM based target classification using RCS feature vectors. / Bufler, Travis Dale; Narayanan, Ram Mohan; Dogaru, Traian.

Radar Sensor Technology XIX; and Active and Passive Signatures VI. ed. / Armin Doerry; Chadwick Todd Hawley; G. Charmaine Gilbreath; Kenneth I. Ranney. SPIE, 2015. 94610I (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 9461).

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

TY - GEN

T1 - SVM based target classification using RCS feature vectors

AU - Bufler, Travis Dale

AU - Narayanan, Ram Mohan

AU - Dogaru, Traian

PY - 2015/1/1

Y1 - 2015/1/1

N2 - This paper investigates the application of SVM (Support Vector Machines) for the classification of stationary human targets and indoor clutter via spectral features. Applying Finite Difference Time Domain (FDTD) techniques allows us to examine the radar cross section (RCS) of humans and indoor clutter objects by utilizing different types of computer models. FDTD allows for the spectral characteristics to be acquired over a wide range of frequencies, polarizations, aspect angles, and materials. The acquired target and clutter RCS spectral characteristics are then investigated in terms of their potential for target classification using SVMs. Based upon variables such as frequency and polarization, a SVM classifier can be trained to classify unknown targets as a human or clutter. Furthermore, the application of feature selection is applied to the spectral characteristics to determine the SVM classification accuracy of a reduced dataset. Classification accuracies of nearly 90% are achieved using radial and polynomial kernels.

AB - This paper investigates the application of SVM (Support Vector Machines) for the classification of stationary human targets and indoor clutter via spectral features. Applying Finite Difference Time Domain (FDTD) techniques allows us to examine the radar cross section (RCS) of humans and indoor clutter objects by utilizing different types of computer models. FDTD allows for the spectral characteristics to be acquired over a wide range of frequencies, polarizations, aspect angles, and materials. The acquired target and clutter RCS spectral characteristics are then investigated in terms of their potential for target classification using SVMs. Based upon variables such as frequency and polarization, a SVM classifier can be trained to classify unknown targets as a human or clutter. Furthermore, the application of feature selection is applied to the spectral characteristics to determine the SVM classification accuracy of a reduced dataset. Classification accuracies of nearly 90% are achieved using radial and polynomial kernels.

UR - http://www.scopus.com/inward/record.url?scp=84954052308&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84954052308&partnerID=8YFLogxK

U2 - 10.1117/12.2176759

DO - 10.1117/12.2176759

M3 - Conference contribution

AN - SCOPUS:84954052308

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Radar Sensor Technology XIX; and Active and Passive Signatures VI

A2 - Doerry, Armin

A2 - Hawley, Chadwick Todd

A2 - Gilbreath, G. Charmaine

A2 - Ranney, Kenneth I.

PB - SPIE

ER -

Bufler TD, Narayanan RM, Dogaru T. SVM based target classification using RCS feature vectors. In Doerry A, Hawley CT, Gilbreath GC, Ranney KI, editors, Radar Sensor Technology XIX; and Active and Passive Signatures VI. SPIE. 2015. 94610I. (Proceedings of SPIE - The International Society for Optical Engineering). https://doi.org/10.1117/12.2176759