Invisible Optical Adversarial Stripes on Traffic Sign against Autonomous Vehicles

TL;DR

This paper introduces GhostStripe, a stealthy optical attack using LED stripes that exploits rolling shutter effects to mislead autonomous vehicle traffic sign recognition, demonstrating high stability and potential safety risks.

Contribution

The paper presents GhostStripe, a novel attack system that creates invisible adversarial stripes on traffic signs by controlling LED light timing to exploit rolling shutter effects.

Findings

Achieves up to 94% misclassification rate in real testbeds

Stealthy attack invisible to humans but effective against cameras

Potential to cause life-threatening incidents in autonomous driving

Abstract

Camera-based computer vision is essential to autonomous vehicle's perception. This paper 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 human. 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. We discuss the countermeasures at the levels of camera sensor, perception model, and autonomous driving system.