English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Adversarial Examples (AEs) can deceive Deep Neural Networks (DNNs) and havereceived a lot of attention recently. However, majority of the research on AEsis in the digital domain and the adversarial patches are static, which is verydifferent from many real-world DNN applications such as Traffic SignRecognition (TSR) systems in autonomous vehicles. In TSR systems, objectdetectors use DNNs to process streaming video in real time. From the view ofobject detectors, the traffic sign`s position and quality of the video arecontinuously changing, rendering the digital AEs ineffective in the physicalworld.  In this paper, we propose a systematic pipeline to generate robust physicalAEs against real-world object detectors. Robustness is achieved in three ways.First, we simulate the in-vehicle cameras by extending the distribution ofimage transformations with the blur transformation and the resolutiontransformation. Second, we design the single and multiple bounding boxesfilters to improve the efficiency of the perturbation training. Third, weconsider four representative attack vectors, namely Hiding Attack, AppearanceAttack, Non-Target Attack and Target Attack.  We perform a comprehensive set of experiments under a variety ofenvironmental conditions, and considering illuminations in sunny and cloudyweather as well as at night. The experimental results show that the physicalAEs generated from our pipeline are effective and robust when attacking theYOLO v5 based TSR system. The attacks have good transferability and can deceiveother state-of-the-art object detectors. We launched HA and NTA on a brand-new2021 model vehicle. Both attacks are successful in fooling the TSR system,which could be a life-threatening case for autonomous vehicles. Finally, wediscuss three defense mechanisms based on image preprocessing, AEs detection,and model enhancing.

Fooling the Eyes of Autonomous Vehicles: Robust Physical Adversarial Examples Against Traffic Sign Recognition Systems