T-Mobile 5G Is Linking Wildfire-Detecting AI Cameras To Put Out Fires Faster

T-Mobile has partnered with the startup Pano AI to spot wildfires before they get out of control. CNET reports: The startup Pano AI uses a series of cameras that survey the wilderness and AI algorithms that watch for telltale smoke -- an indicator of small blazes that could grow into raging wildfires. That footage is sent to the startup's headquarters for human confirmation, and if a fire is burning, evidence is sent to clients who could be affected. While Pano AI had been sending evidence photos over 4G LTE networks at slow rates of around 20 to 30 6-megapixel images per minute, its new partnership with T-Mobile has it using the carrier's 5G network to send video at 30 frames per second, which is around 90 times more data. Ultimately, getting evidence to Pano AI's clients, which include utility companies, much quicker on 5G means a faster response from firefighters and potentially squashing big fires before they get dangerous.

Pano AI works with a number of utilities, governments, fire authorities, forestry companies and private landlords who in turn work with local emergency responders. Its newest client and the first with a system using T-Mobile's 5G network is Portland General Electric (PGE), a utility supplying gas and electricity to 16 million customers around Portland, Oregon. Pano AI has 20 cameras set up in the forests surrounding the city that give 10-mile panoramic views, which include powerlines. This lets PGE know if fires are headed toward its infrastructure.

T-Mobile recruited Pano AI to be part of its Innovation Lab alongside other companies harnessing 5G to improve their services, such as Mixhalo, which is using the carrier's 5G network to pipe in concert audio directly to audience members' phones. But Pano AI's partnership goes deeper, as it's mounting its cameras on T-Mobile's cell towers, saving months of time and paperwork needed to request and install its equipment on other signal towers or similar vantage points.

Nice app. But...

[Ungrounded Lightning]

Nice app.

But the infrastructure involves cameras watching a lot of territory, linked in real time to an AI that can recognize the signs of a fire.

What else could such an AI recognize? Once the video and feeds are in place, just add some more recognition (maybe clone the feeds to another crunch farm) and you've got the panopticon.

What could possibly go wrong?

Re:

[AmiMoJo]

If they are using "AI" it's probably very simple. They may well just be using an infrared detector that is tuned to trigger when something is on fire. Similar to presence detection but tuned for much higher temperatures.

Main reason being that you want remote devices to be extremely low power, so that they can run indefinitely from solar power. IR detectors don't use much power. Alternatively capture say one frame per minute and look for red pixels.

Re:

[Narcocide]

How that AI source code looks:

if(burning == true){
    print "Aak! I'm on fire!!"
}

Re:

[chill]

You jest, but it is probably just a git pull away from this [wikipedia.org].

Re:

[tlhIngan]

Yeah, these devices are going to be around for a decade or more, and to get to them is probably someone hiking for a week to get there, so they need to literally be completely autonomous in operation.

They will have a small solar cell (too big and it's a problem in the wind) a large battery and very low power circuitry so the solar cell can keep the battery charged and power the circuits. The high-powered stuff like 5G and cameras remain off as keeping them on will drain the battery.

So they will likely have

Re:Nice app. But...

[thegarbz]

What could possibly go wrong?

Man are you in for a surprise when you learn about firewatch towers and the government paid employees who sit in them with binoculars. Hint: Masturbate in your own home not in the forest.

Re:

[sarren1901]

This!!! I watched a video about these firewatch towers in remote parts of the grand canyon and else where. Though I thought a lot of the jobs are going away due to technology and frankly it's hard to find young people willing to go off grid completely. Cell coverage isn't exactly great in the middle of no where.

I imagine a lot of the towers use radios and only the regional branch would have a wired connection to civilization. Eventually you get a signal XD

Re:

[PPH]

Cell coverage isn't exactly great in the middle of no where.

So, how do they expect these AI cameras to work? Hint: Fires most often start and grow out of control in the more rural areas.

Re:

[Junta]

With the caveat that in fact, a number of companies do care who you are, what you do, and where you go, toward the end of manipulating your spend.

Of course, none of those parties would care about the details of your hiking/camping activities, it's enough for their purposes to know that you do some hiking/camping to try to sell you relevant product.

Re: Nice app. But...

[Åke Malmgren]

A number of government agencies also care about collecting as much information as possible on everyone, so that if someone turns out to be interesting in retrospect, they already have a bunch of stuff to go on. Example: https://arstechnica.com/tech-p... [arstechnica.com]

Re:

[sarren1901]

So you are so weak of will that you can't say no to a voluntary transaction?

I must be some demi-god because advertising really only works when I already want something anyway and then you offer a deal, yada. It is no crime, nor even a manipulation, to provide a product or service that someone already wants or could find a use for.

If I didn't already block most ads, it would certainly be nice to be advertised to about things I could generally find an application for.

Just wait for them to beam the ads straigh

Re:

[Junta]

I didn't propose any value judgement on targeted ads, just pointing out that 'nobody cares about what you do enough to monitor your activities' is actually false even for the random person.

Re:

[PPH]

Nobody really cares who you are, what you do, or where you go.

As long as you are a prole, know your place and don't try to rise above it. The Party doesn't care about what the proles think or do.

There's a proverb: The nail that sticks out gets hammered down.

Don't stick out and you'll be fine.

Re:

[Junta]

Precisely because they want to cover a lot of territory and their requirements are fairly coarse, this won't be suitable for significant surveillance. We are probably talking about sporadic cameras covering lots of forest with relatively low resolution. They may be able to know about campsites based on smoke from the campfire, but that's about it. It doesn't take much to spot signs of a fire, but it would take a whole lot more to spy on people under the forest canopy in any significant way.

Re:

[suss]

It's for your "security", citizen. Stop asking questions.

I find some of these claims to be hard to believe.

[sabbede]

Are we really supposed to accept the claim that fire department response times are frame-rate dependent? According to this summary, the maximum difference between identifying a fire over 4G vs 5 is exactly three (3) seconds.

So, this is just a publicity puff-piece for T-Mobile. I wonder how much they paid CNET for it.

Re:

[timeOday]

I can't quite tell from the summary, but are they saying before they were getting 20-30 still images per second total from the network of 20 cameras, vs. now full-motion video from each? I could imagine billowing smoke is easier to recognize when you can see it move, vs. being a patch of fog or dust. But still, yeah, it's a puff piece.

I have noticed my cell service gradually improving over the last 5 years though, both coverage and speed. It's hard to say why because with each step such as 5G when it

Profound and makes you think

[GeekWithAKnife]

...what other amazing things T-Mobile's 5G network can do.

I wish they had included a sign up page. Now I have to browse all the way to their website for a good deal on their great 5G!

Re:

[suss]

In cooperation with Pano. Pano Pano Pano [slashvertisement] Pano Pano Pano Pano.
 
/Pano.

Putting out fires is the problem

[140Mandak262Jamuna]

The solution is not to put the fires out. Its to stop developers from putting subdivisions right smack dab in the middle of combustible materials that have accumulated for 100 years by aggressive relentless putting out of fires.

Re:

[140Mandak262Jamuna]

It is not govt's job to put out fires in the forest. As long as it does people will build homes there and demand govt to bail them out.

Re:

[sarren1901]

You forgot all the state money we spend protecting the coast line and all those houses along it.

I guess bandwidth is infinite, then?

[argStyopa]

30 frames per second is nice, but is it really necessary for fire surveillance video?
A forest fire identification isn't going to come in the comparison of frames 0.03 seconds apart.

That's like 5 megabits of pipe to watch trees grow 24/7.

Population of Oregon

[douglasfir77]

16 million customers??? The entire state of Oregon only has 4.19 million. Guessing its supposed to be 1.6 million PGE customers. Also DLR in Germany has been doing something similar for years to watch for forest fires mainly in the forested areas around big cities that still have UXO from WW I and II that like to pop off in the summers. There was a USA company called Fire Watch nearly 10 years ago trying to bring this DLR technology to the USA.

I only read the headline

[burtosis]

So this proves it then, all this time I’d been saying 5g controls the weather and no one would listen. Wake up sheeple, there not going to stop just using rain to put out fires.

Re:

[sarren1901]

I mean, if we could just put up 5g towers in spots of the forest that could detect fire then launch drones that seed clouds that cause it to rain. All automated. Sounds pretty awesome!!!

Though the seeding of the clouds probably has a ram up time that the fire would surely outpace as fire tends to spread pretty rapidly and sometimes spontaneously.

We'll have to radio for the water-dropping planes that will be controlled by AI instead.