Capturing images of bystanders by zooming in on pictures of corneas

The reflections off of the human eye are well studied. http://en.wikipedia.org/wiki/Purkinje_images

It is claimed they can be seen in the image of Our Lady of Guadeloupe http://web.archive.org/web/20071209175102/http://www.interlupe.com.mx/8-e.html

Anyone else remembering that scene from High Anxiety, in which that photographer blows up a photo to a ridiculous size to prove that Mel Brooks’s character was being impersonated?

1 Like

Great; you’ve just invented a market for matte contact lenses. Probably unnecessary given the points made above, but it’ll sell.

On the other hand… what about the work on wavefront cameras that let you select focus point and depth of field after the picture is taken? Getting that AND high pixel resolution may be a challenge, but conceptually…

On the other other hand, a quick reminder: Anything happening in a public place has no reasonable expectation of privacy, whether recorded directly or indirectly. So many, perhaps most, of the cases where this could be used would NOT technically be further erosion of privacy.

1 Like

not exactly a cell phone camera.

Subjects were photographed from a viewing distance of approximately 1 m using a Hasselblad H2D 39 megapixel digital camera (50 ISO; f8 aperture; 1/250 sec. shutter; single shot, manual focus) with 120 mm macro lens. The room was flash illuminated by two Bowens DX1000 lamps with dish reflectors, positioned side by side approximately 80 cm behind the camera, and directed upwards to exclude catch light. Two additional DX1000 flash lamps with soft boxes were positioned behind baffles on either side of the subject to illuminate the bystanders.

quite a difference from your usual cell phone camera. Although a cell phone camera might not be all that more sensitive than ISO 50…

1 Like

Uncrop.

1 Like

https://www.youtube.com/watch?v=sp77AjBdlEc On a related note. :slight_smile:

1 Like

Of course, we can’t omit the postmortem sketches of Father Tubelcek’s eyes, in Terry Pratchett’s Feet of Clay Discworld novel, revealing the reflected glow of a Golem’s eyes:

Then he took his iconograph out of its box and prepared to take a picture of the corpse. As he did so, something caught his eye. Father Tubelcek lay there, one eye still open as Vimes had left it, winking at eternity. Cheery looked closer. He’d thought he’d imagined it. But … Even now he wasn’t sure. The mind could play tricks.

He opened the little door of the iconograph and spoke to the imp inside. ‘Can you paint a picture of his eye, Sydney?’ he said. The imp squinted out through the lens. ‘Just the eye?’ it squeaked. ‘Yes. As big as you can.’ […]
‘Bigger. So big it fills the whole paper. In fact’ - Cheery squinted at the picture in his hands - ‘just paint the pupil. The bit in the middle.’ […] Another damp picture unwound. It showed a big black disc. Well… mainly black. Cheery looked closer. There was a hint, just a hint … He rapped on the box again. ‘Yes, Mr Dwarf Weird Person?’ said the imp. ‘The bit in the middle. Big as you can, thank you.’ […]

And the paper was black … except for the tiny little area that wasn’t.

and then…

‘Er … have you ever heard the story about dead men’s eyes, sir?’

‘Assume I haven’t had a literary education, Littlebottom.’

‘Well… they say …’ […] ‘They say that the last thing a dying man sees stays imprinted in his eyes, sir.’

‘Oh, that. That’s just an old story.’

‘Yes. Amazing, really. I mean, if it weren’t true, you’d have thought it wouldn’t have survived, wouldn’t you? I thought I saw this little red spark, so I got the imp to paint a really big picture before it faded completely. And, right in the centre …’

2 Likes

Now that you mention it… I remember the big hype about wavefront cameras but nothing else. Anything actually on the market right now?

I’ve seen an end cap display at Best Buy. I think it was fairly recently.
Lytro, if I rightly remember.

Lytro had a pre-Xmas sale so I picked one up. Still experimenting with it, I see the potential for it but, despite software upgrades, it feels like a Version 1.0. Interestingly, some of my experiments capture reflections in my cat’s eye, but the depth of field is not sufficient to glean useful data other than the binary “the sun is out”.

(and other, related, “I can’t work out how to do this, therefore it is impossible” type posts)

Technology only ever gets better. Consider what astronomers have done and are doing to ‘untangle’ images from outer space which are blurred by imperfections in our atmosphere - literally taking the twinkle out of stars. Consider also what another group of astronomers did to post-correct the massive flaw in the Hubble Space Telescopes mirror; turning smudged nonsense into some of the most aesthetically beautiful and scientifically important photos ever taken.

Making the first atomic weapons was the work of thousands and took a significant fraction of the world’s physics PhDs. Now it’s the work of mere technicians, and the only hurdle is actually getting your hands on a few of the specific bits, which is just a matter of logistics.

Can this be done with current, common technology? No.
Is it a concern in the next 5-10 years? No, probably not.
Is it something my kids will have to concern themselves with at some point in their life? Now the genie is out of the box in the real world (as opposed to CSI, or some such), I’d say the answer to that is definitely yes.

2 Likes

I didn’t say that it wouldn’t work at all. It is not that hard to look for eye photographs in flickr and be able to se the photographer in the reflection. I just don’t expect this to be a very practical way of identifying people because of the fundamental way light behaves. I’ve done this kind of thing before. It is tricky as all hell to get right angle, focus, etc. and a small change in conditions has significant effects.

Look at these two images. They were taken about 30 seconds apart

http://imgur.com/iDyUTAl,dsEBUWa

Notice how in the first one the reflection is only apart of the cat’s eye and on the second one it fills it. What happened was my cat moved her head a tiny bit, mere milimeters. Ontop of that, in order to photograph that reflection with out overexposing the rest of the frame I had to put a chair over her for shade, make sure she was looking to where the sun was shining and I was about a centimeter or so away from her face. It takes quite a bit of work to get these conditions and they don’t happen everywhere, and the smallest thing (like a cloud overhead) can mess them up.

What you are fighting against is regular and diffractive optics effects that have been known about since the mid 1600’s and math worked out since the 1800’s. What the researchers did is neat, but they setup the conditions before hand and worked with the person being photograph to make certain the reflection in his eye got photographed.

If you think you can identify someone by the eyes of a portrait of someone else with something the size of a cell phone camera I wish you the best of luck because airy disks ( http://en.wikipedia.org/wiki/Airy_disk ) are going to get in your way really fast.

1 Like

Not comparable today, yes.

Not comparable 5,10 years from now?

Actually, this is being used to help capture and prosecute hands on pedophiles (you know, the ones that do the raping and taping - the ones that rip out anuses and vaginas of babies). I am okay with that.

Sensor sizes will always matter—regardless of any increase in the light-capturing efficiency of sensors—because of diffraction limitations, as soimless has alluded to above…

This is news? I give you the “enhance” scene from the movie “The Blade Runner”.

Theoretically, the diffraction limit can be overcome

This topic was automatically closed after 5 days. New replies are no longer allowed.