Spoofing Mobileye 630’s Video Camera Using a Projector

Nassi, Ben; Nassi, Dudi; Ben-Netanel, Raz; Elovici, Yuval

doi:10.14722/autosec.2021.23007

Cited by 6 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results demonstrated that the misclassification of the altered indicators led to certain unanticipated actions in the car. In a different study [213] that focused on the wellknown Mobileye external ADAS, researchers used drone projections of fake traffic signals to determine how environmental changes-such as those related to projection speed, diameter, color, and ambient light-affected the success of an attack.…”

Section: Ai-based Adversarial Attacks Against Autonomous Vehiclesmentioning

confidence: 99%

Cybersecurity of Autonomous Vehicles: A Systematic Literature Review of Adversarial Attacks and Defense Models

Girdhar

Hong

Moore

2023

IEEE Open J. Veh. Technol.

View full text Add to dashboard Cite

Autonomous driving (AD) has developed tremendously in parallel with the ongoing development and improvement of deep learning (DL) technology. However, the uptake of artificial intelligence (AI) in AD as the core enabling technology raises serious cybersecurity issues. An enhanced attack surface has been spurred on by the rising digitization of vehicles and the integration of AI features. The performance of the autonomous vehicle (AV)-based applications is constrained by the DL models' susceptibility to adversarial attacks despite their great potential. Hence, AI-enabled AVs face numerous security threats, which prevent the large-scale adoption of AVs. Therefore, it becomes crucial to evolve existing cybersecurity practices to deal with risks associated with the increased uptake of AI. Furthermore, putting defense models into practice against adversarial attacks has grown in importance as a field of study amongst researchers. Therefore, this study seeks to provide an overview of the most recent adversarial defensive and attack models developed in the domain of AD.

show abstract

Section: Ai-based Adversarial Attacks Against Autonomous Vehiclesmentioning

confidence: 99%

Cybersecurity of Autonomous Vehicles: A Systematic Literature Review of Adversarial Attacks and Defense Models

Girdhar

Hong

Moore

2023

IEEE Open J. Veh. Technol.

View full text Add to dashboard Cite

show abstract

“…The use of computer vision systems in AVs has become a field of relevance in cybersecurity for its potential impact on the safety of citizen. Almost all perception tasks (e.g., sign and object recognition, localisation or object tracking) performed in AVs are currently based on deep neural network architectures [25,28,29,36,48,52]. All these systems are susceptible to evasion attacks and other vulnerabilities specific to such systems [2,14,47,50].…”

Section: Experimental Evaluation Of the Feasibility Of A Physical Adv...mentioning

confidence: 99%

Exploring the Feasibility of Physical Adversarial Attacks: A Cybersecurity Study

Hamon,

Junklewitz

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Adversarial machine learning (AML), by designing attacks that intentionally break or misuse state-of-the-art machine learning models, has become the most prominent scientific field to explore the security aspects of Artificial Intelligence. A whole range of vulnerabilities, previously irrelevant in traditional ICT, have effectively emerged in these studies. In the light of upcoming legislations mandating security requirements for AI products and services, there is a need to understand how AML techniques connect with the broader field of cybersecurity, and how to articulate more tightly threat models with realistic cybersecurity procedures. This article aims to contribute to closing the gap between AML and cybersecurity by proposing an approach to study the feasibility of an attack in a cybersecurity risk assessment framework, illustrated with a specific use case of an evasion attack designed to fool traffic sign recognition systems in the physical world. The importance of considering the feasibility of carrying out such attacks under real conditions is emphasized through the analysis of two factors: the reproducibility of the attack according to a published description or existing code, and the applicability of the attack by a malicious actor operating in a real-world environment.

show abstract

“…Without this capability, Tesla's autopilot was unintentionally triggered, causing the car to: (1) continuously slam on the brakes in response to a print advertisement containing a picture of a person that appeared on a bus [4], and (2) stop in response to a billboard advertisement that contained a stop sign [5]. Moreover, attackers can exploit the absence of this capability and intentionally trigger: (1) Tesla's autopilot to suddenly stop the car in the middle of a road in response to a stop sign embedded in an advertisement on a digital billboard [2], and (2) Mobileye 630 to issue false notifications regarding a projected road sign [6].…”

Section: Related Workmentioning

confidence: 99%

EyeDAS: Securing Perception of Autonomous Cars Against the Stereoblindness Syndrome

Levy¹,

Nassi²,

Swissa³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The ability to detect whether an object is a 2D or 3D object is extremely important in autonomous driving, since a detection error can have life-threatening consequences, endangering the safety of the driver, passengers, pedestrians, and others on the road. Methods proposed to distinguish between 2 and 3D objects (e.g., liveness detection methods) are not suitable for autonomous driving, because they are object dependent or do not consider the constraints associated with autonomous driving (e.g., the need for real-time decision-making while the vehicle is moving). In this paper, we present EyeDAS, a novel few-shot learning-based method aimed at securing an object detector (OD) against the threat posed by the stereoblindness syndrome (i.e., the inability to distinguish between 2D and 3D objects). We evaluate EyeDAS's real-time performance using 2,000 objects extracted from seven YouTube video recordings of street views taken by a dash cam from the driver's seat perspective. When applying EyeDAS to seven state-of-the-art ODs as a countermeasure, EyeDAS was able to reduce the 2D misclassification rate from 71.42-100% to 2.4% with a 3D misclassification rate of 0% (TPR of 1.0). We also show that EyeDAS outperforms the baseline method and achieves an AUC of over 0.999 and a TPR of 1.0 with an FPR of 0.024.

show abstract

Spoofing Mobileye 630’s Video Camera Using a Projector

Cited by 6 publications

References 6 publications

Cybersecurity of Autonomous Vehicles: A Systematic Literature Review of Adversarial Attacks and Defense Models

Cybersecurity of Autonomous Vehicles: A Systematic Literature Review of Adversarial Attacks and Defense Models

Exploring the Feasibility of Physical Adversarial Attacks: A Cybersecurity Study

EyeDAS: Securing Perception of Autonomous Cars Against the Stereoblindness Syndrome

Contact Info

Product

Resources

About