A recursive kinematic random forest and alpha beta filter classifier for 2D radar tracks

Lars W. Jochumsen, J. Østergaard, Søren H. Jensen, Carmine Clemente, Morten Ø Pedersen

Research output: Contribution to journalArticle

108 Downloads (Pure)

Abstract

In this work, we show that by using a recursive random forest together with an alpha beta filter classifier it is possible to classify radar tracks from the tracks’ kinematic data. The kinematic data is from a 2D scanning radar without Doppler or height information. We use random forest as this classifier implicit handles the uncertainty in the position measurements. As stationary targets can have an apparently high speed because of the measurement uncertainty, we use an alpha beta filter classifier to classify stationary targets from moving targets. We show an overall classification rate from simulated data at 82.6 % and from real world data 79.7 %. Additional to the confusion matrix we also show recordings of real world data.
Original languageEnglish
Number of pages17
JournalEURASIP Journal on Advances in Signal Processing
Publication statusAccepted/In press - 5 Jul 2016

Fingerprint

Kinematics
Radar
Classifiers
Doppler radar
Position measurement
Scanning
Uncertainty

Keywords

  • radar
  • classification
  • random forest
  • alpha beta filter
  • kinematic

Cite this

@article{ce4ec815c8ce45688636e2d1148d1dc0,
title = "A recursive kinematic random forest and alpha beta filter classifier for 2D radar tracks",
abstract = "In this work, we show that by using a recursive random forest together with an alpha beta filter classifier it is possible to classify radar tracks from the tracks’ kinematic data. The kinematic data is from a 2D scanning radar without Doppler or height information. We use random forest as this classifier implicit handles the uncertainty in the position measurements. As stationary targets can have an apparently high speed because of the measurement uncertainty, we use an alpha beta filter classifier to classify stationary targets from moving targets. We show an overall classification rate from simulated data at 82.6 {\%} and from real world data 79.7 {\%}. Additional to the confusion matrix we also show recordings of real world data.",
keywords = "radar, classification, random forest, alpha beta filter, kinematic",
author = "Jochumsen, {Lars W.} and J. {\O}stergaard and Jensen, {S{\o}ren H.} and Carmine Clemente and Pedersen, {Morten {\O}}",
year = "2016",
month = "7",
day = "5",
language = "English",
journal = "EURASIP Journal on Advances in Signal Processing",
issn = "1110-8657",

}

A recursive kinematic random forest and alpha beta filter classifier for 2D radar tracks. / Jochumsen, Lars W.; Østergaard, J.; Jensen, Søren H.; Clemente, Carmine; Pedersen, Morten Ø.

In: EURASIP Journal on Advances in Signal Processing, 05.07.2016.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A recursive kinematic random forest and alpha beta filter classifier for 2D radar tracks

AU - Jochumsen, Lars W.

AU - Østergaard, J.

AU - Jensen, Søren H.

AU - Clemente, Carmine

AU - Pedersen, Morten Ø

PY - 2016/7/5

Y1 - 2016/7/5

N2 - In this work, we show that by using a recursive random forest together with an alpha beta filter classifier it is possible to classify radar tracks from the tracks’ kinematic data. The kinematic data is from a 2D scanning radar without Doppler or height information. We use random forest as this classifier implicit handles the uncertainty in the position measurements. As stationary targets can have an apparently high speed because of the measurement uncertainty, we use an alpha beta filter classifier to classify stationary targets from moving targets. We show an overall classification rate from simulated data at 82.6 % and from real world data 79.7 %. Additional to the confusion matrix we also show recordings of real world data.

AB - In this work, we show that by using a recursive random forest together with an alpha beta filter classifier it is possible to classify radar tracks from the tracks’ kinematic data. The kinematic data is from a 2D scanning radar without Doppler or height information. We use random forest as this classifier implicit handles the uncertainty in the position measurements. As stationary targets can have an apparently high speed because of the measurement uncertainty, we use an alpha beta filter classifier to classify stationary targets from moving targets. We show an overall classification rate from simulated data at 82.6 % and from real world data 79.7 %. Additional to the confusion matrix we also show recordings of real world data.

KW - radar

KW - classification

KW - random forest

KW - alpha beta filter

KW - kinematic

UR - http://asp.eurasipjournals.springeropen.com/

M3 - Article

JO - EURASIP Journal on Advances in Signal Processing

JF - EURASIP Journal on Advances in Signal Processing

SN - 1110-8657

ER -