[HN Gopher] Insect-worn microcamera streams video to phone 120 m...
___________________________________________________________________
Insect-worn microcamera streams video to phone 120 meters away
Author : bookofjoe
Score : 213 points
Date : 2020-07-19 16:35 UTC (6 hours ago)
(HTM) web link (spectrum.ieee.org)
(TXT) w3m dump (spectrum.ieee.org)
| lifeisstillgood wrote:
| I'be been banging on about MOOP - Massive open online psychology-
| where we basically monitor our daily interactions such as
| conversations with kids partners etc, and build a society wide
| set of best practises and can then be guided in real time by
| these best practises (if you have ever seen those shows where a
| Nanny lives full time for a week with a family that the kind of
| thing)
|
| But it only works if the data is treated _medically_. we won 't
| ever stop the three letter agencies from abusing it (entirely)
| but we need to make PII more than GDPR and make it as sacrosanct
| in law as lawyer client / doctor patient confidentiality.
| fearingreprisal wrote:
| Seek professional psychological help.
| ourmandave wrote:
| I suppose the spooks at the CIA are all, "5fps, you're adorable."
| mchan889 wrote:
| "That's a large bug, have you considered something smaller?
| Maybe a mosquito?"
|
| - Someone somewhere on a DARPA project
| tomrod wrote:
| Fascinating! But I am a bit disappointed there isn't a video
| taken solely by an insect. All the tech highlights without a
| demo.
| DoofusOfDeath wrote:
| Agreed. It's really bizarre that they didn't include even one
| frame of the transmitted video. It's central to the story.
| fwip wrote:
| It's in the video embedded in the page.
| jameslk wrote:
| Doesn't the video in the article show this? Here's the video:
| https://youtu.be/115BGUZopHs
| ummonk wrote:
| I think people overlook it (I did) because it looks like an
| ad.
| SMAAART wrote:
| Holy Batman! Now apply Moore's law to that and figure out where
| we'd be in 10-20 years #SCARY_AF
| arthurcolle wrote:
| Moore's Law doesn't really work anymore. It hasn't held for the
| last 4 years, IIRC. That's why a lot of focus has been on
| multicore systems to try and get new speedups based on parallel
| computing as opposed to getting purely single-chip
| improvements.
| mpfundstein wrote:
| it indeed doesn't work anymore for single CPUs but still
| seems to hold for other thibgs. eg memory (ddr5) or chipsize
| (see 7nm chips)
| ComputerGuru wrote:
| Moores law is strictly about transistor count and not
| performance, it is just that we are only now seeing the two
| come decoupled.
| beamatronic wrote:
| Also figure the US Government probably had this 5 years ago
| op03 wrote:
| I'd like a spider cam. I find it a great mystery how spiders
| survive in rooms which seem absolutely empty/abandoned etc.
| throwanem wrote:
| Often they can survive quite a while without food or water,
| as long as the temperature and humidity are within some
| fairly generous extremes, although very dry air poses the
| risk of dehydration since their respiratory systems and body
| volume offer little ability to retain a reserve. But they
| rarely have to go _that_ long without a meal, because if a
| spider can get in, so can suitably sized prey - and will.
| Next time you find yourself wondering how a spider thrives in
| an empty room, take a close look at and under her web. Most
| likely you 'll find cocooned prey in both places - fresher
| prey items in the web, spent ones underneath it.
|
| Spinning a web is a metabolically very expensive process,
| requiring as it does a great deal of protein - protein which
| can only be recovered by either eating the spun silk again,
| or eating prey trapped by it. Spiders consequently have to be
| pretty smart about where and how they do it, because the ones
| who aren't tend not to reproduce. So it's usually a very safe
| bet that, wherever you see a spider web, it's there because
| that's where the prey tends to be. It can be interesting to
| see what sort of prey that is - for example, right now in the
| basement of my ~130-year-old house there are a couple of very
| happy cellar spiders enjoying a steady diet of pill bugs, and
| another by the kitchen radiator who all on her own solved
| most of an ant problem for me before I ever realized I had
| it.
|
| They're actually very desirable animals to have around, in my
| thoroughly considered opinion, despite getting about as bad a
| name as wasps - and with just as little reason.
| lanius wrote:
| Ditto for cockroaches. I'm curious where those little buggers
| disappear to when they manage to escape from me.
| amelius wrote:
| I'm thinking of
| https://en.wikipedia.org/wiki/Monkey_selfie_copyright_disput...
|
| Who will own the copyright on the video?
| james1966 wrote:
| They used a rather large camera...
|
| The smallest known camera module is the OmniVision OV6948, has
| the dimensions: 0.65 x 0.65 x 1.158mm, including 120 degree wide
| lens.
|
| Sadly it cannot be purchased in single quantities :(
|
| [1] https://www.digitalcameraworld.com/news/worlds-smallest-
| came...
| sevencolors wrote:
| The furthest depth of field is only 30mm so it can't see very
| far
| contingencies wrote:
| Could you use an array of these mounted at precise, known,
| differing angles and an FPGA to calculate high resolution
| localised depth maps? Applications you say? Entomological
| gesture studies!
| bredren wrote:
| Sometimes if you write the manufacturer you can get sample
| amounts. They are priced much higher than normal but I've done
| this with LED components.
|
| Edit: It may help to describe your sample order request as
| being part of an evaluation.
| tofof wrote:
| Interesting. That OV06948 camera is also 40k (200x200) pixels,
| 30fps, and RGB. Critically, the page you link lists its weight
| at 0.87g = 870mg, which would make it 157x denser than lead.[1]
| The actual spec sheet says 0.87 mg, i.e. less than one
| milligram, which is incredible.[2] It has up to a 4 meter
| range, which I understand to be wired, so it would still need
| some sort of transmitter.
|
| The camera used by these researchers, including its lens and
| panning head, is 200 mg (77 for the sensor and lens and mount,
| 7 mg bluetooth chip, and 96 mg boost converter). It's only 20k
| (160x120) resolution, monochrome. It has a wireless 120 m
| range.
|
| The killer, however, seems to be power. The 10 mAh battery
| alone weighs 500 mg, which means the 200 mg vs <1 mg comparison
| becomes more like 700 vs 501.
|
| The OV06948's power usage is 25 mW, which means a 10 mAh / (25
| mW / 3.3 V) = 1.32 hour runtime, capturing 142k frames of
| video. The bug rig gets 6 hours, capturing only 108,000 frames
| of video. Perhaps you could get longer battery life out of the
| OV06948 if you slowed the framerate.
|
| [1]
| https://www.wolframalpha.com/input/?i=%28870+mg+%2F+%280.65+...
|
| [2]
| https://www.ovt.com/download/sensorpdf/553/OmniVision_OVM694...
| zaroth wrote:
| Unfortunately this seems outside the bounds for ambient RF
| energy harvesting, even if you don't mind dropping a few
| frames to charge back up if the capacitor runs low.
|
| The rectenna would have to be much too large to produce
| anywhere near 25mW even at a 10% duty cycle.
|
| I suppose if you didn't mind ramping the RF output power way
| up you could shrink the antenna size, but not much smaller
| than 1cm^2 I think.
| apocalypstyx wrote:
| And yet once again it's demonstrated we're living Philip K.
| Dick's "Reality".
| rv-de wrote:
| Am I the only one here at least considering the ethical aspect of
| attaching some device to an insect?
|
| Maybe I'm sentimental and contradictory because when I drive a
| car over the Autobahn I'm killing hundreds of insects in an hour.
| At the same time I cannot harm a bug intentionally.
| 6gvONxR4sf7o wrote:
| It's like killing a spider in the house. Killing it instead of
| taking it outside doesn't bring me any benefit the way eating a
| chicken does, so why kill it? Just like putting the camera rig
| on an actual beetle for the demo doesn't bring a benefit, so
| why do it?
| cgriswald wrote:
| That's only true if your time has no value. Killing is
| faster. Also, as someone who has gotten multiple infections
| from non-venomous spider bites, I can tell you catch and
| release is riskier than killing.
| matonias wrote:
| I think this is just a reflection of your personality.
| Respecting nature, small or big and taking the time to give
| it space, let's you value and organise your life in a more
| balanced fashion.
| stormdennis wrote:
| Spiders live by the sword so when I see them in my bath, I am
| become God!
| istjohn wrote:
| > The researchers removed the electronics from the insects
| after the experiments and observed no noticeable adverse
| effects on their behavior.
| mycall wrote:
| You don't have to drive.
| toxicFork wrote:
| someone has to kill the insects
| Sharlin wrote:
| Attaching a camera to an insect is probably not even a blip on
| the radar of a university ethics board compared to all the
| mammals that are used and killed annually in various
| experiments.
| raz32dust wrote:
| I can relate, but considering how much we exploit animals,
| including agricultural animals, breeding, raising and killing
| animals for food, using rats and monkeys for testing etc., this
| would not surface in the priority list for fighting animal
| welfare for me for a very, very long time. Even in thought.
| LeifCarrotson wrote:
| I agree - there's a big difference between the ability of
| various species to experience suffering or to experience
| happiness, and I think that has an impact on the ethics of
| how we treat them. Humans, monkeys, pigs, dogs, etc. are
| definitely social/emotional intelligences that deserve better
| treatment than we give them. Rats? Chickens? Insects?
| Jellyfish? Plants? Bacteria? Somewhere on that spectrum I
| cease to worry about the suffering of the animal and worry
| more about those affected by it.
| FiberBundle wrote:
| That's so ignorant. How can we possibly know how other
| animals experience pain or suffering? You just assume that
| the further you diverge from us in the animal kingdom the
| less animals are able to suffer. That certainly makes
| things easier.
| ChuckMcM wrote:
| Excellent work. I'm old enough to remember when putting wireless
| vision systems on a robot required a kilogram at least :-). Of
| course this would be made even simpler if you integrated the
| vision sensor, the BLE radio, and the control system into a
| single die. Something that would be quite accessible to nation
| states and sufficiently bored billionaires. Ambient energy
| harvesting has also made great strides so the days of persistent
| and difficult detect surveillance are getting closer.
| fearingreprisal wrote:
| Here's technology that certainly couldn't possibly be misused to
| infringe upon civil liberties...
| TLightful wrote:
| Great. I'm a huge nature lover. Literally wouldn't hurt a fly.
|
| But I'll be stocking up on insect-killing sprays.
|
| To coin a phrase, this is the "progress of d!ckheads".
| woodruffw wrote:
| Yet another win for the Fifth Element[1].
|
| [1]: https://www.youtube.com/watch?v=DrHMBletjXg
| rs23296008n1 wrote:
| That shoe sums up my response to remote controlled roach
| cameras.
| barmstrong wrote:
| Is there a list of others?
|
| I love using sci fi as inspiration for new products.
| tijuco2 wrote:
| I've seen another study, which in my opinion is way more useful,
| where they are able to control a cockroaches using electrical
| pulses on their antennas. With a camera attached to its back,
| these cockroachs will be able to find people underneath rubbles.
| hilbertseries wrote:
| How deep into rubble could it get before losing signal?
| mirekrusin wrote:
| They could do p2p.
| hiccuphippo wrote:
| They could call it a swarm.
| yellowapple wrote:
| Sounds like it'd be pretty buggy, though.
| EGreg wrote:
| Please send a link
| tijuco2 wrote:
| https://youtu.be/icQv33kWKm8
| emrehan wrote:
| Kickstarter, 2013:
| https://www.kickstarter.com/projects/backyardbrains/the-
| robo...
| Nbox9 wrote:
| Could a cockroach carry both the cockroach control device and
| the camera? Does the micro camera have enough low light
| capabilities to be useful in the dark? Are there any animal
| ethics to be taken into consideration for attempting this
| route?
|
| I can't tell if that idea is 2 years or 20 years in the future,
| but it sounds neat.
| wongarsu wrote:
| You'd probably be better off using a microphone if we are
| taking about search in low light conditions. Or maybe a CO2
| sensor, if you can miniaturize it enough.
|
| But I'm not so sure it's that much better than dogs for
| search and rescue. Much more promising for surveillance.
| papito wrote:
| Pickle Riiiiiick!
| saagarjha wrote:
| I've heard that such roaches have a short working life, though,
| because at some point they'll begin ignoring antenna input.
| loktarogar wrote:
| In Fifth Element, pretty much exactly this was used to spy on
| the President. (it was squished)
| ramblerman wrote:
| Bah, this really is the low hanging fruit of progress.
|
| Make cameras smaller, and super glue them to a bug. Is there
| really a point, besides the trendy headline?
| JulianMorrison wrote:
| Even spies have hobbies.
|
| You might be able to do some interesting wildlife documentary
| movies, though.
| quickthrower2 wrote:
| It's possible to remote control a cockroach sized bot, so now
| you can be a sort of "fly on the wall" in any conversation.
| Drop the roach off by drone which can double up as the access
| point.
| 1123581321 wrote:
| Miniaturization and weight reduction of existing technology are
| worthy scientific pursuits.
| walterbell wrote:
| From the article:
|
| > _The power system is the primary limitation here, but it might
| be possible to use a solar cell to cut down on battery
| requirements ... with a long-range wireless link and a vision
| system, it's possible to add sophisticated vision-based autonomy
| to tiny robots by doing the computation remotely_
|
| Since WiFi can be used to see through walls and ceilings of
| buildings and homes, using low-cost passive sensors, such devices
| can be autonomously navigated. E.g. for indoor use, knowing light
| source locations for charging and camouflage.
|
| From
| https://www.sciencedirect.com/science/article/pii/S235234092...
| (2020):
|
| _> our dataset provides a collection of Wi-Fi signals that are
| recorded for 40 different pairs of subjects while performing
| twelve two-person interactions. The presented dataset can be
| exploited to advance Wi-Fi-based human activity recognition in
| different aspects, such as the use of various machine learning
| algorithms to recognize different human-to-human interactions._
|
| Related project: https://dhalperi.github.io/linux-80211n-csitool/
| OriPekelman wrote:
| OK kids, I need help here. There was a sci-fi novel, probably the
| sixties, called something like "little bugs have little bugs on
| them". The premise was basically that.. A cold war where one side
| puts bugs on the bugs of the other side. Rings any bells?
| Potentially there were also some aspects of bug driven
| assassinations... But I am unsure. This thing been bugging me.
| ficklepickle wrote:
| I think I may have found it on the wikipedia page for
| nanotechnology in fiction:
|
| > The 1984 novel Peace on Earth by Stanislaw Lem tells about
| small bacteria-sized nanorobots looking as normal dust
| (developed by artificial intelligence placed by humans on the
| Moon in the era of cold warfare) that has later came to Earth
| and are replicating, destroying all weapons, modern technology
| and software, leaving living organisms (as there were no living
| organisms on the Moon) intact.
|
| It's not a perfect match for your description, nor is it from
| the 60s, but it was the closest match I could find. Is that the
| book you were thinking of?
| matonias wrote:
| Sounds like Corona
| partiallypro wrote:
| The resolution is so low it's not of much use for intel services
| at the moment (160x120 monochrome.)
| jahnu wrote:
| Are you kidding?
| ur-whale wrote:
| Unless they've managed to somehow hook the output of the cam to
| the cortex of the bug thereby allowing it to do things it
| previously couldn't, I find the article a tad disappointing: this
| is one _huge_ bug. Does it even fly?
| sillysaurusx wrote:
| The title made it sound like a fly or an ant, but in reality it's
| a rather large beetle. It's neat work though.
|
| For a redteam project, I once made a spy camera that fit in a
| belt buckle. I'm not sure you'll need to worry about surveillance
| implications of insect-mounted cameras; surveillance is already
| pervasive.
| Nbox9 wrote:
| Imagine releasing the bug near a restricted area. The attacker
| can hope it gets past security and gain images from inside the
| restricted area. If it gets caught then the attacker if up to
| 120M away, which is a good head start.
| echelon wrote:
| Or the video is bounced over a cell-enabled relay and the
| attacker is nowhere nearby.
| Nbox9 wrote:
| Good point. The maximum distance the attacker could be
| would be limited by the act of releasing an insect and the
| biological limitations of insects (lifespan, travel speed),
| not by the range of the wireless communication.
| koliber wrote:
| Which can be mounted on larger rodents and also sent loose
| in the area.
| mhb wrote:
| Could use a repeater
| jameslk wrote:
| Cool tech but I'm not as excited about the surveillance this will
| enable
| throwanem wrote:
| I'd worry a lot less about any theoretical Project Electric
| Cockroach than I would about ubiquitous smartphones.
| zokier wrote:
| I don't think camera system size has been limiting factor in
| surveillance for a while. For robotics (either walking or
| flying) the locomotion and power are more restricting aspects,
| and using live insects for surveillance seems too slow and
| unpredictable to be useful.
|
| Check out Black Hornet Nano for what was state of art almost a
| decade ago https://en.wikipedia.org/wiki/Black_Hornet_Nano
| m0xte wrote:
| At least you can step on it :)
| Mirioron wrote:
| Then you get sued for destroying government property.
| throwanem wrote:
| And admit the program's existence in open court? No, either
| nothing happens and you sound like a loon, or you get
| black-bagged and just disappear.
|
| ...a joke that hits _way different_ after hearing the news
| out of Portland...
| johnwheeler wrote:
| The proverbial "fly on the wall"
| mhh__ wrote:
| If it's remotely possible I can almost guarantee the three
| letter agencies around the world are using it already (or have
| tried).
|
| MI5 were lifting encryption keys via acoustic analysis of
| typing something like 65 years ago [Wright, Spycatcher]. And
| their budget was fairly limited, as opposed to the US agencies
| with their basically unlimited budgets (i.e. NRO)
| crispyporkbites wrote:
| Once they get it down to flying insects, would it be possible to
| deploy a swarm of these (thousands) and use positioning data to
| generate high resolution 3D imagery?
| thegreekgonzo wrote:
| Or jumping insects?
|
| https://en.wikipedia.org/wiki/Spotted_lanternfly
| bookofjoe wrote:
| "Prey" -- Michael Crichton (2002)
|
| https://en.wikipedia.org/wiki/Prey_(novel)
| thysultan wrote:
| Hopefully the smaller it gets the less power it needs, the
| easier it is to power using environmental factors like wind,
| earths magnetic fields and the movement of the insect in such a
| scenario.
| Nbox9 wrote:
| It would be a very hard challenging to build a high resolution
| image from several low level images. I haven't heard of any
| similar work. This is an interesting problem space.
| nikhizzle wrote:
| What you are asking for is along the lines of Microsoft
| Photosynth released circa 2008.
|
| https://en.wikipedia.org/wiki/Photosynth
| wintermutestwin wrote:
| That app was awesome and it's functionality found its way
| into the Microsoft Pix app.
| contingencies wrote:
| The academic field is called
| https://en.wikipedia.org/wiki/Photogrammetry and was an
| outgrowth of early military aviation. There are now numerous
| open source implementations.
___________________________________________________________________
(page generated 2020-07-19 23:00 UTC)