Year: 2025
Type: article
Abstract: Camera-based computer vision is essential to autonomous vehicle’s perception. This article presents an attack that uses light-emitting diodes and exploits the camera’s rolling shutter effect to create adversarial stripes in the captured images to mislead traffic sign recognition. The attack is stealthy because the stripes on the traffic sign are invisible to humans. For the attack to be threatening, the recognition results need to be stable over consecutive image frames. To achieve this, we design and implement GhostStripe , an attack system that controls the timing of the modulated light emission to adapt to camera operations and victim vehicle movements. Evaluated on real testbeds, GhostStripe can stably spoof the traffic sign recognition results for up to 94% of frames to a wrong class when the victim vehicle passes the road section. In reality, such attack effect may fool victim vehicles into life-threatening incidents. To counteract this threat, we propose GhostBuster , a software-based defense module to detect and mitigate the effects of GhostStripe. GhostBuster incorporates a perturbation detector and a sign restorer, effectively restoring the natural appearance of compromised traffic signs and significantly reducing the attack’s impact. We also discuss other countermeasures at the levels of camera sensor and autonomous driving system.
DOI: 10.1145/3773031
Authors: Dongfang Guo, Yuting Wu, Pengfei Zhou, Xin Lou, Rui Tan
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Award: G5622044399
OpenAlex ID: W7101427834