Convolutional Neural Networks for Robust Classification of Drones

Holly Dale; Mohammed Jahangir; Christopher J. Baker; Michail Antoniou; Stephen Harman; Bashar I. Ahmad

doi:10.1109/RadarConf2248738.2022.9764172

Convolutional Neural Networks for Robust Classification of Drones

Holly Dale, Mohammed Jahangir, Christopher J. Baker, Michail Antoniou, Stephen Harman, Bashar I. Ahmad

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

Abstract

In order to be effective, radar drone surveillance systems need to be able to discriminate between birds and drones. In this work, convolutional neural networks (CNNs) are used to distinguish between bird and drone spectrograms, where the classifier is tested on real, low signal to background ratio (SBR) data obtained using an L-band staring radar. This allows for a better understanding of the classifier's ability to generalise against new models of drone and new clutter environments. This work highlights the importance of SBR for drone surveillance, placing limits on the size of drone that can be reliably classified, as well as range from the radar.

Original language	English
Title of host publication	2022 IEEE Radar Conference (RadarConf22)
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781728153681
ISBN (Print)	9781728153698 (PoD)
DOIs	https://doi.org/10.1109/RadarConf2248738.2022.9764172
Publication status	Published - 3 May 2022
Event	2022 IEEE Radar Conference, RadarConf 2022 - New York City, United States Duration: 21 Mar 2022 → 25 Mar 2022

Publication series

Name	Proceedings of the IEEE Radar Conference
Publisher	IEEE
ISSN (Print)	1097-5764
ISSN (Electronic)	2640-7736

Conference

Conference	2022 IEEE Radar Conference, RadarConf 2022
Country/Territory	United States
City	New York City
Period	21/03/22 → 25/03/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

classification
convolutional neural networks
staring radar
UAVs

ASJC Scopus subject areas

Computer Networks and Communications
Signal Processing
Instrumentation

Access to Document

10.1109/RadarConf2248738.2022.9764172

Cite this

@inproceedings{aa91a53384114a758f09a8c505797b94,

title = "Convolutional Neural Networks for Robust Classification of Drones",

abstract = "In order to be effective, radar drone surveillance systems need to be able to discriminate between birds and drones. In this work, convolutional neural networks (CNNs) are used to distinguish between bird and drone spectrograms, where the classifier is tested on real, low signal to background ratio (SBR) data obtained using an L-band staring radar. This allows for a better understanding of the classifier's ability to generalise against new models of drone and new clutter environments. This work highlights the importance of SBR for drone surveillance, placing limits on the size of drone that can be reliably classified, as well as range from the radar.",

keywords = "classification, convolutional neural networks, staring radar, UAVs",

author = "Holly Dale and Mohammed Jahangir and Baker, {Christopher J.} and Michail Antoniou and Stephen Harman and Ahmad, {Bashar I.}",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE Radar Conference, RadarConf 2022 ; Conference date: 21-03-2022 Through 25-03-2022",

year = "2022",

month = may,

day = "3",

doi = "10.1109/RadarConf2248738.2022.9764172",

language = "English",

isbn = "9781728153698 (PoD)",

series = "Proceedings of the IEEE Radar Conference",

publisher = "IEEE",

booktitle = "2022 IEEE Radar Conference (RadarConf22)",

}

Dale, H, Jahangir, M, Baker, CJ, Antoniou, M, Harman, S & Ahmad, BI 2022, Convolutional Neural Networks for Robust Classification of Drones. in 2022 IEEE Radar Conference (RadarConf22)., 9764172, Proceedings of the IEEE Radar Conference, IEEE, 2022 IEEE Radar Conference, RadarConf 2022, New York City, United States, 21/03/22. https://doi.org/10.1109/RadarConf2248738.2022.9764172

TY - GEN

T1 - Convolutional Neural Networks for Robust Classification of Drones

AU - Dale, Holly

AU - Jahangir, Mohammed

AU - Baker, Christopher J.

AU - Antoniou, Michail

AU - Harman, Stephen

AU - Ahmad, Bashar I.

PY - 2022/5/3

Y1 - 2022/5/3

N2 - In order to be effective, radar drone surveillance systems need to be able to discriminate between birds and drones. In this work, convolutional neural networks (CNNs) are used to distinguish between bird and drone spectrograms, where the classifier is tested on real, low signal to background ratio (SBR) data obtained using an L-band staring radar. This allows for a better understanding of the classifier's ability to generalise against new models of drone and new clutter environments. This work highlights the importance of SBR for drone surveillance, placing limits on the size of drone that can be reliably classified, as well as range from the radar.

AB - In order to be effective, radar drone surveillance systems need to be able to discriminate between birds and drones. In this work, convolutional neural networks (CNNs) are used to distinguish between bird and drone spectrograms, where the classifier is tested on real, low signal to background ratio (SBR) data obtained using an L-band staring radar. This allows for a better understanding of the classifier's ability to generalise against new models of drone and new clutter environments. This work highlights the importance of SBR for drone surveillance, placing limits on the size of drone that can be reliably classified, as well as range from the radar.

KW - classification

KW - convolutional neural networks

KW - staring radar

KW - UAVs

UR - http://www.scopus.com/inward/record.url?scp=85146197161&partnerID=8YFLogxK

U2 - 10.1109/RadarConf2248738.2022.9764172

DO - 10.1109/RadarConf2248738.2022.9764172

M3 - Conference contribution

AN - SCOPUS:85146197161

SN - 9781728153698 (PoD)

T3 - Proceedings of the IEEE Radar Conference

BT - 2022 IEEE Radar Conference (RadarConf22)

PB - IEEE

T2 - 2022 IEEE Radar Conference, RadarConf 2022

Y2 - 21 March 2022 through 25 March 2022

ER -

Convolutional Neural Networks for Robust Classification of Drones

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this