The researchers suggest redesigning driverless car cameras so they scan rows of pixels in a random order, rather than one row at a time, would make it harder for an attacker to introduce color interference.
Credit: temp-64GTX/Shutterstock
Researchers at China's Zhejiang University found driverless cars could be fooled into seeing red traffic lights as green.
The scientists directed a laser at the sensors of five camera models used by self-driving vehicles, with two open-source software packages reading the captured images.
Lasers of a 650-nanometer and a 520-nanometer wavelength rendered the entire image red or green, respectively, while flickering the laser at high frequencies only induced this coloration in certain image segments.
Adding a horizontal bar of green or red caused both software packages to incorrectly sense the traffic lights as green 30% of the time and red 86% of the time, on average, across the cameras.
From New Scientist
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA
No entries found