deflock.me

552 readers

333 users here now

Automated License Plate Readers (ALPRs) are AI-powered cameras that capture and analyze images of all passing vehicles, storing details like your car’s location, date, and time. They also capture your car’s make, model, color, and identifying features such as dents, roof racks, and bumper stickers, often turning these into searchable data points. These cameras collect data on millions of vehicles—regardless of whether the driver is suspected of a crime. While these systems can be useful for tracking stolen cars or wanted individuals, they are mostly used to track the movements of innocent people.

Learn more at deflock.me

founded 11 months ago

MODERATORS

cm0002@sh.itjust.works

cm0002@infosec.pub

Can Flock cameras and similar camera networks be tricked with projected images/video? (lemmy.world)

submitted 2 days ago by GamingChairModel@lemmy.world to c/DeFlock@sh.itjust.works

5 comments fedilink hide all child comments

In recent years, I've seen people use projectors at night to protest policies: projecting slogans, images, etc., using a normal projector against the wall of the target of a protest (embassies, government buildings, prisons/detainment facilities, corporate headquarters, etc.).

Here's an example, but I've seen lots of other examples. And even outside of protest, people have been doing ridiculous things with 3D projection, like this guy animating a hotel cuck chair (SFW).

This got me to thinking: if the image recognition of Flock and similar cameras is just an AI image recognition function (using ImageNet or something like that) being run on captured video, what's stopping people from projecting false data onto streets and sidewalks in a way that tricks the Flock camera into recording false people/vehicles in places where there was no person/vehicle? Including fake license plates, etc.?

Seems like an attack vector for some kind of denial of service attack, like putting fake faces and license plates all over.

Is there anyone doing this? Would it be possible, at least at night, for certain camera locations and angles?

top 5 comments

sorted by: hot top controversial new old

[–] OakTree@lemmy.zip 1 points 8 hours ago

It sounds like flock also does some sort of tracking with TPMS sensors. Basically new tires wirelessly report back to your cars computer, and that signal is unique enough to track. There might be room to flood them with fake TPMS data

[–] honesthenery@thelemmy.club -1 points 1 day ago

[–] dr_yeti@lemmy.world 5 points 2 days ago* (last edited 2 days ago) (1 children)

I don't know about projection, but adversarial patterns on cars and clothes are one way to trick these systems. Benn Jordan has done some work on plate decals that make cars difficult to parse by plate readers (eg Youtube video). Bill Swearingen is developing adversarial clothing (eg his kickstarter). 'Adversarial pattern' or 'adversarial clothing' are good search terms if you're interested.

Edit: These guys are working with a heat-sensitive patch that powers up in a minute, then provides adversarial patterning in visible and infra-red. That's maybe closer to what you had in mind.

[–] pammyfromthesticks@mander.xyz -1 points 2 days ago

Anything but over throw your goverment

[–] Chozo@fedia.io 3 points 2 days ago

I don't know that Flock cameras are using any depth-sensing hardware, so it stands to reason that they could be defeated with 2D projections.

One issue that you run into, however, is that these cameras are typically pointed at places where people are. That means that you'd have to be projecting these images onto places where people are going to be. This could be disorientating on pedestrian areas, and could be catastrophic on roadways.