An image-based approach for classification of human micro-doppler radar signatures

Fok Hing Chi Tivive; Son Lam Phung; Abdesselam Bouzerdoum

doi:10.1117/12.2015707

An image-based approach for classification of human micro-doppler radar signatures

Fok Hing Chi Tivive^*, Son Lam Phung, Abdesselam Bouzerdoum

^*Corresponding author for this work

University of Wollongong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

15 Citations (Scopus)

Abstract

With the advances in radar technology, there is an increasing interest in automatic radar-based human gait identification. This is because radar signals can penetrate through most dielectric materials. In this paper, an image-based approach is proposed for classifying human micro-Doppler radar signatures. The time-varying radar signal is first converted into a time-frequency representation, which is then cast as a two-dimensional image. A descriptor is developed to extract micro-Doppler features from local time-frequency patches centered along the torso Doppler frequency. Experimental results based on real data collected from a 24-GHz Doppler radar showed that the proposed approach achieves promising classification performance.

Original language	English
Title of host publication	Active and Passive Signatures IV
DOIs	https://doi.org/10.1117/12.2015707
Publication status	Published - 2013
Externally published	Yes
Event	Active and Passive Signatures IV - Baltimore, MD, United States Duration: 1 May 2013 → 2 May 2013

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	8734
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Active and Passive Signatures IV
Country/Territory	United States
City	Baltimore, MD
Period	1/05/13 → 2/05/13

Keywords

Human micro-Doppler radar signature
Log-Gabor filters
Micro-Doppler descriptor
Spectrogram
Support vector machines
Two-directional two-dimensional principal component analysis

Access to Document

10.1117/12.2015707

Cite this

@inproceedings{b51b45e49dcc48ab868673a47e588436,

title = "An image-based approach for classification of human micro-doppler radar signatures",

abstract = "With the advances in radar technology, there is an increasing interest in automatic radar-based human gait identification. This is because radar signals can penetrate through most dielectric materials. In this paper, an image-based approach is proposed for classifying human micro-Doppler radar signatures. The time-varying radar signal is first converted into a time-frequency representation, which is then cast as a two-dimensional image. A descriptor is developed to extract micro-Doppler features from local time-frequency patches centered along the torso Doppler frequency. Experimental results based on real data collected from a 24-GHz Doppler radar showed that the proposed approach achieves promising classification performance.",

keywords = "Human micro-Doppler radar signature, Log-Gabor filters, Micro-Doppler descriptor, Spectrogram, Support vector machines, Two-directional two-dimensional principal component analysis",

author = "Tivive, {Fok Hing Chi} and Phung, {Son Lam} and Abdesselam Bouzerdoum",

year = "2013",

doi = "10.1117/12.2015707",

language = "English",

isbn = "9780819495259",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

booktitle = "Active and Passive Signatures IV",

note = "Active and Passive Signatures IV ; Conference date: 01-05-2013 Through 02-05-2013",

}

An image-based approach for classification of human micro-doppler radar signatures. / Tivive, Fok Hing Chi; Phung, Son Lam; Bouzerdoum, Abdesselam.
Active and Passive Signatures IV. 2013. 873406 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 8734).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An image-based approach for classification of human micro-doppler radar signatures

AU - Tivive, Fok Hing Chi

AU - Phung, Son Lam

AU - Bouzerdoum, Abdesselam

PY - 2013

Y1 - 2013

N2 - With the advances in radar technology, there is an increasing interest in automatic radar-based human gait identification. This is because radar signals can penetrate through most dielectric materials. In this paper, an image-based approach is proposed for classifying human micro-Doppler radar signatures. The time-varying radar signal is first converted into a time-frequency representation, which is then cast as a two-dimensional image. A descriptor is developed to extract micro-Doppler features from local time-frequency patches centered along the torso Doppler frequency. Experimental results based on real data collected from a 24-GHz Doppler radar showed that the proposed approach achieves promising classification performance.

AB - With the advances in radar technology, there is an increasing interest in automatic radar-based human gait identification. This is because radar signals can penetrate through most dielectric materials. In this paper, an image-based approach is proposed for classifying human micro-Doppler radar signatures. The time-varying radar signal is first converted into a time-frequency representation, which is then cast as a two-dimensional image. A descriptor is developed to extract micro-Doppler features from local time-frequency patches centered along the torso Doppler frequency. Experimental results based on real data collected from a 24-GHz Doppler radar showed that the proposed approach achieves promising classification performance.

KW - Human micro-Doppler radar signature

KW - Log-Gabor filters

KW - Micro-Doppler descriptor

KW - Spectrogram

KW - Support vector machines

KW - Two-directional two-dimensional principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=84881172154&partnerID=8YFLogxK

U2 - 10.1117/12.2015707

DO - 10.1117/12.2015707

M3 - Conference contribution

AN - SCOPUS:84881172154

SN - 9780819495259

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

BT - Active and Passive Signatures IV

T2 - Active and Passive Signatures IV

Y2 - 1 May 2013 through 2 May 2013

ER -

An image-based approach for classification of human micro-doppler radar signatures

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this