Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems

Abdul Jabbar Siddiqui; Azzedine Boukerche

doi:10.1145/3416013.3426461

Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems

Abdul Jabbar Siddiqui
, Azzedine Boukerche

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

4 Scopus citations

Abstract

The recent works on automated vehicle make and model recognition (VMMR) have embraced the use of advanced deep learning models such as convolutional neural networks. In this work, we introduce an adversarial attack against such VMMR systems through adversarially learnt patches. We demonstrate the effectiveness of the adversarial patches against VMMR through experimental evaluations on a real-world surveillance dataset. The developed adversarial patches achieve reductions of upto 37% in VMMR recall scores. It is hoped that this work shall motivate future studies in developing VMMR systems that are robust to adversarial learning-based attacks.

Original language	English
Title of host publication	Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks
Publisher	Association for Computing Machinery, Inc
Pages	115-121
Number of pages	7
ISBN (Electronic)	9781450381208
DOIs	https://doi.org/10.1145/3416013.3426461
State	Published - 16 Nov 2020
Externally published	Yes

Publication series

Name	Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks

Bibliographical note

Publisher Copyright:
© 2020 ACM.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities

Keywords

adversarial attacks
adversarial patches
adversarial robustness
cyber-physical systems security
smart city surveillance
vehicle make and model recognition

ASJC Scopus subject areas

Computer Networks and Communications
Software
Electrical and Electronic Engineering
Safety, Risk, Reliability and Quality

Access to Document

10.1145/3416013.3426461

Cite this

Siddiqui, A. J., & Boukerche, A. (2020). Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems. In Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks (pp. 115-121). (Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks). Association for Computing Machinery, Inc. https://doi.org/10.1145/3416013.3426461

Siddiqui, Abdul Jabbar ; Boukerche, Azzedine. / Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems. Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks. Association for Computing Machinery, Inc, 2020. pp. 115-121 (Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks).

@inproceedings{489207be852d40c8b814843f00a40227,

title = "Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems",

abstract = "The recent works on automated vehicle make and model recognition (VMMR) have embraced the use of advanced deep learning models such as convolutional neural networks. In this work, we introduce an adversarial attack against such VMMR systems through adversarially learnt patches. We demonstrate the effectiveness of the adversarial patches against VMMR through experimental evaluations on a real-world surveillance dataset. The developed adversarial patches achieve reductions of upto 37\% in VMMR recall scores. It is hoped that this work shall motivate future studies in developing VMMR systems that are robust to adversarial learning-based attacks.",

keywords = "adversarial attacks, adversarial patches, adversarial robustness, cyber-physical systems security, smart city surveillance, vehicle make and model recognition",

author = "Siddiqui, \{Abdul Jabbar\} and Azzedine Boukerche",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.",

year = "2020",

month = nov,

day = "16",

doi = "10.1145/3416013.3426461",

language = "English",

series = "Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks",

publisher = "Association for Computing Machinery, Inc",

pages = "115--121",

booktitle = "Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks",

}

Siddiqui, AJ & Boukerche, A 2020, Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems. in Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks. Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks, Association for Computing Machinery, Inc, pp. 115-121. https://doi.org/10.1145/3416013.3426461

Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems. / Siddiqui, Abdul Jabbar; Boukerche, Azzedine.
Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks. Association for Computing Machinery, Inc, 2020. p. 115-121 (Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks).

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

TY - GEN

T1 - Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems

AU - Siddiqui, Abdul Jabbar

AU - Boukerche, Azzedine

PY - 2020/11/16

Y1 - 2020/11/16

N2 - The recent works on automated vehicle make and model recognition (VMMR) have embraced the use of advanced deep learning models such as convolutional neural networks. In this work, we introduce an adversarial attack against such VMMR systems through adversarially learnt patches. We demonstrate the effectiveness of the adversarial patches against VMMR through experimental evaluations on a real-world surveillance dataset. The developed adversarial patches achieve reductions of upto 37% in VMMR recall scores. It is hoped that this work shall motivate future studies in developing VMMR systems that are robust to adversarial learning-based attacks.

AB - The recent works on automated vehicle make and model recognition (VMMR) have embraced the use of advanced deep learning models such as convolutional neural networks. In this work, we introduce an adversarial attack against such VMMR systems through adversarially learnt patches. We demonstrate the effectiveness of the adversarial patches against VMMR through experimental evaluations on a real-world surveillance dataset. The developed adversarial patches achieve reductions of upto 37% in VMMR recall scores. It is hoped that this work shall motivate future studies in developing VMMR systems that are robust to adversarial learning-based attacks.

KW - adversarial attacks

KW - adversarial patches

KW - adversarial robustness

KW - cyber-physical systems security

KW - smart city surveillance

KW - vehicle make and model recognition

UR - https://www.scopus.com/pages/publications/85097647239

U2 - 10.1145/3416013.3426461

DO - 10.1145/3416013.3426461

M3 - Conference contribution

AN - SCOPUS:85097647239

T3 - Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks

SP - 115

EP - 121

BT - Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks

PB - Association for Computing Machinery, Inc

ER -

Siddiqui AJ, Boukerche A. Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems. In Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks. Association for Computing Machinery, Inc. 2020. p. 115-121. (Q2SWinet 2020 - Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks). doi: 10.1145/3416013.3426461

Adversarial Patches-based Attacks on Automated Vehicle Make and Model Recognition Systems

Abstract

Publication series

Bibliographical note

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this