SoylentNews is people
SoylentNews
Touting the technology as a replacement for IDs, part of your laptop's login, or able to let cops know if the person they just pulled over is dangerous, the first effective long-range iris scanner has been developed by Marios Savvides, a Carnegie Mellon engineering professor:
"Fingerprints, they require you to touch something. Iris, we can capture it at a distance, so we're making the whole user experience much less intrusive, much more comfortable," Savvides [said]. Unlike other scanners, which required someone to step up to a machine, his scanner can capture someone's iris and face as they walk by.
"There's no X-marks-the-spot. There's no place you have to stand. Anywhere between six and 12 meters, it will find you, it will zoom in and capture both irises and full face," he said.
Iris scanning currently works only at close range, so it requires a level of cooperation of the person being scanned:
"It requires a level of cooperation that makes it very overt—a person knows that you're taking a picture for this purpose,"...If it succeeds, long-distance scanning will change all that. Savvides says his team has secured a patent for his invention and will continue to work to make it easier and cheaper. He continues, too, to look for positive implementations of it.
Spotted on The Eponymous Pickle.
(Score: 4, Informative) by Nerdfest on Friday May 15 2015, @01:26AM
I hate to be the bearer of bad news, but when researching iris scanners as part of a job, the manufacturers were already supplying scanners with a range of 100+ metres to 'certain customers' and apparently they had quite a good acquire rate. Even 5 years ago I was shown scanners for airports, etc that were picking up scans from lines of people as they walked by at about 4-6 metres. Perhaps I'm missing something, but this stuff has already been done, and apparently done better.
(Score: 3, Insightful) by captain normal on Friday May 15 2015, @04:42AM
If these work so damn well why do we have to line up and put all our stuff through x-ray machines, take off our shoes and walk through a magnetic coil? If they know who we are, they should know whether or not we're a terrorist.
When life isn't going right, go left.
(Score: 2) by Yog-Yogguth on Sunday May 17 2015, @12:49AM
The following turned into a long and slightly off-topic comment, sorry about that.
… they should know whether or not we're a terrorist.
On so many different levels that's not how it works and I'll focus on one specific one that I think will become more apparent over time: you're either a confirmed terrorist or a potential terrorist, there's not in any meaningful way any such thing as a non-terrorist because humans aren't static nor particularly stable.
Anything can turn a potential terrorist into an actual terrorist, a thought, a serendipitous moment, a new experience, a change in their environment, a change of definitions, manipulation, subversion, trickery. It can be minute. They might not even know it themselves.
Like many other people (millions) I'm most likely already classified as an extremist since among plenty of other things I'm interested enough in Linux to have looked at the websites of Tails and the Linux Journal and who knows what else they flag you for when they flag that. It's unlikely to do me any favors that I have military weapons experience.
Even if they could read your thoughts they couldn't feel certain (it's not like your brain is “read only”), there will always be real or imaginary loopholes to fear (and they're getting there, have a look at the practical civilian commercial brain interface [wikipedia.org] accessories for computer use like Emotiv [wikipedia.org]). It's not the sensory part that is challenging and lacking so far but the correct interpretation of the data.
Feel free to start watching (and thinking about) things like Ghost in the Shell and maybe Psycho-Pass, and read William Gibson, Neal Stephenson, Vernor Vinge, John Brunner, Philip K. Dick. etc. and snap up the cultural themes of popular/iconic games like Half Life, Halo, etc. (did you know the “triple-A” FPS gaming industry now has some links to military development and especially brainstorming?). It's not that they're necessarily identical but a lot of the issues raised have parallel examples that aren't all that far fetched any more.
Maybe it makes the point if I ask whether or not you believe hypnosis [wikipedia.org] is real?
Unknown unknowns are also impossible to check for.
The surveillance and manipulation system as it exists has one very big “flaw”: currently (or more precisely: last we knew i.e. when Snowden left more than a year ago) for most things someone has/had to search for the information that is already there and analyze it, manually.
The system itself as it existed then had only relatively limited powers to bring things to attention as far as we know (there is too little detail about the project/codename MONSTER and similar but we do know plenty about Snowden's job as an analyst and can infer from that).
They're obviously working hard at expanding that; it's the whole point of and idea behind big data and machine learning to do precisely that as a tool for distilling actionable “knowledge”/information.
And it's unlikely to be a single underfunded guy in a basement working on it by himself. One year is half a computer “generation”, Moore's Law didn't stop last I checked.
Will it be ten years until everybody takes constant scanning for granted? Who knows. I don't think enough people have realized what we're heading for yet. I doubt any of the legislation is going to do much but at least some of the people “in power” in the US government has started to become a tiny bit worried since they've reportedly tried to get a better view of what is actually going on.
Bite harder Ouroboros, bite! tails.boum.org/ linux USB CD secure desktop IRC *crypt tor (not endorsements (XKeyScore))
(Score: 2) by Yog-Yogguth on Sunday May 17 2015, @12:55AM
Correction: I meant MONSTERMIND.
Bite harder Ouroboros, bite! tails.boum.org/ linux USB CD secure desktop IRC *crypt tor (not endorsements (XKeyScore))
(Score: 0) by Anonymous Coward on Friday May 15 2015, @04:18PM
As a guy who makes his living working with, and designing, optical systems from sensors to optics, I have great skepticism that this system will work, or that any in the past have worked. There are always plenty of devices touted that never pan out. I recall from some 20 years ago there was much ado about a sensor that could detect someone's heartbeat through walls from many meters away. Lots of units were sold to groups like the Special Ops, largely because it was backed by a Senator from Pennsylvania. The company that made them demonstrated how well they worked, but the demonstrations were far from being realistic nor "double-blind." In practice, it never really worked (but it wasn't much of a surprise to scientists and engineers who thought about how it was supposed to work).
100+ meters, and on a non-cooperative target? Let's assume some conservative numbers. Take a person fresh from an eye exam, whose pupils are dialated out to 10 mm (and that would be like, creepy, anime eyes). To image that, you need to have about three camera pixels to cover that, so the pixels need to cover about 3 mm in object space. Then, to make out features on the retina, you need to resolve an order of magnitude better than that, so a pixel would need to cover, say, 300 microns in object space, from a distance of 100 meters. Now, even if we assume the sensor has a pixel size of only 1-micron, this means the sensor needs an imager on it with a focal length of (1-micron)*(100m)/(300-micron) = 333mm optic. Not unreasonable. But, look at all the assumptions made, make them realistic, and you're taking multi-meter focal length optics (or more). I'm not buying it.
(Score: 0) by Anonymous Coward on Saturday May 16 2015, @06:13PM
yeah, yeah also some dudes sold dowsing rods that can detect explosives, unfortunately these were assigned to real troops instead of airport security theatres.