Slashdot Mirror
Algorithm Reveals Objects Hidden Behind Other Things In Camera Phone Images
KentuckyFC writes "Imaging is undergoing a quiet revolution at the moment thanks to various new techniques for extracting data from images. Now physicists have worked out how to create an image of an object hidden behind a translucent material using little more than an ordinary smartphone and some clever data processing. The team placed objects behind materials that scatter light such as onion skin, frosted glass and chicken breast tissue. They photographed them using a Nokia Lumina 1020 smartphone, with a 41 megapixel sensor. To the naked eye, the resulting images look like random speckle. But by treating the data from each pixel separately and looking for correlations between pixels, the team was able to produce images of the hidden objects. They even photographed light scattered off a white wall and recovered an image of the reflected scene--a technique that effectively looks round corners. The new technique has applications in areas such as surveillance and medical imaging."
"So save your "enhance!" jokes."
pan left, grid 54
"So save yo
zoom grid 54 27, pan right
"So d save yo
pan right
"So don't ......... save yo
stop, pan left, hold
"So don't save your "enhance!" jokes."
*Vangelis starts playing*
You mean like the frosted glass commonly used for bathroom windows and shower doors? I see this as being a form of image processing that will rapidly be perfected.
-- This and all my posts are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
It is object behind a translucent screen. It does some AI based image sharpening. It is not the classic tank behind a tree. Nor reconstructing a face partially hidden by faces or hoodies etc.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Tits or it never happened.
I am sold!
The government will be using this to spy on you.
Are you still sold?
A feeling of having made the same mistake before: Deja Foobar
But this is far more entertaining: http://media.moronail.net/imag...
Yes, I enjoy the thought of people getting paid to look at my naked flabby body. Their disgust turns me on. I really love hitting on my enhanced patdown specialists. I fly twice a week, and it's the thing I look forward to most in life right now. I'd tip them if it wouldn't put me on the no-fly list. The best part is when they run the back of their hand over my groin and find "something rigid" and then need to double check.
Some of us are interested in the theory, not to mention that both the article and the paper contain actual images. It's not the fault of the authors that you didn't bother to read them through.
This is an impressive step forward in image processing - while reconstructing an image from diffuse light seemed plausible in theory, figuring out how to do it in practice is a hard problem. These guys deserve some respect.
I read the summary.
They had me at "breast".
"Flyin' in just a sweet place,
Never been known to fail..."
First thing I thought of, also. "X-ray glasses" as an app for your phone. Um, great...
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
"a technique that effectively looks round corners"
What does that mean? Obviously you meant "a technique that effectively looks around corners" but you were too lazy.
Not only that but you can get sued by Apple for using round corners.
No, you see all of the images from the paper right up at the top. The kicker is that it can only do monochrome. So your x-ray voyeur shots will only work for girls and boys wearing full-body paint under their clothes. Or maybe you can illuminate them with lasers. I'm sure nobody will notice.
An algorithm for perceiving objects hidden behind other objects could...enable even men to find things in the refrigerator!
How could you possibly do this experiment without trying it through the frosted glass of a shower door with a naked person on the other side?
-Styopa
I don't even want to know the intended practical application of the chicken skin. Just don't tell T-S-A about it :-)
Table-ized A.I.
I am sold!
The government will be using this to spy on you. Are you still sold?
Of course! They're already got access to the home and computer cameras. I pay taxes; why can't I share in the fun as well?
Also, kinda reminds me of that zooming mirror scene in Blade Runner.
If the universe is someone's simulation -- does that mean the stars are just stuck pixels?
Can I get this in a Car analogy?
I guess it would be like driving a Chevy Lumia van.
Yo dawg, I heard you like the Ackermann function, so OH GOD OH GOD OH GOD
I don't claim to be an imaging expert, but a few odd details about the experimental method jumped out at me. It's been known for some time now that diffusive and other scene-perturbing objects (e.g. grossly distorting 'lenses' such as a Coke bottle) can be nullified using a structured light technique to characterize and effectively 'undo' the perturber. A simple structured light example is to replace the light source with a DLP projector and take multiple images with only one pixel illuminated at a time. More clever implementations can replace the single pixels with speckle patterns, zebra stripes, etc., and replace the 2D imager with a single-pixel photocell. Other neat tricks can then be performed such as reconstructing the image from the POV of the light source rather than the imaging device.
The experimentals shown in this paper all seem to have two things in common: 1) the "object" in each case is a backlit, 2D binary pattern on a transparency film or similar, with a relatively small illuminated area, and 2) an extremely narrowband (laser, actually) light source is used. The paper does mention several times that the light source is non-coherent, but it is a laser under the hood. This explains the numerous references to "speckle" in the images, which may leave most readers scratching their heads since things don't normally speckle when looked at through a slice of onion under ordinary light. Speckling is a laser (de)coherence phenomenon where the rays are put slightly out of coherence so as to interfere constructively and destructively.
These things suggest to me that while the paper is definitely interesting, there is no need to worry about the neighbors snapping passable nudes through your shower door or Feds cataloging your grow farm via pictures of a blank wall through your window. This sounds more like a modest extension to what's already been done stirring coherent and structured-light in a pot with convolution and autocorrellation methods.
Since the coherence length of cheap semiconductor lasers (e.g. laser pointers) can be on the order of 1mm or less, it's possible to call even a straight-up laserbeam "non-coherent narrowband light" with a somewhat straight face. Likewise, the quasi-point-sources created using a sparse geometric 2D aperture in transparency film, backlight by the aforementioned source, is pretty close to structured light for practical purposes. The takeaway message is these are very special lighting and "scene" conditions that are not representative of everyday photographic circumstances. So not to worry just yet :-)
Caveat Emptor is not a business model.
Sensor size (in terms of MP) can certainly help, but without a decent lens most of the data is just noise and garbage pixels. A 8MP DLSR with a good lens will give you a vastly better image than a smartphone with a 20+ MP sensor on it.
A big part of getting a good image is how much light is being captured, but you also need a high-quality sensor to capture it accurately. The actual resolution of the sensor is largely a secondary factor in the final image quality. And when you're dealing with a tiny lens on a smartphone, not only do you have focus issues but even the slightest smudge or dust particle will block or distort a very significant portion of the image.
The paper does explain these limitations and it would be wrong to assume that this means we can see/photograph through fog, skin, clouds... But a semiconductor laser can have a spacial coherence long enough to do holography so if they said the source was a laser and non-conherent narroband, I'll give them the benefit of the doubt and assume they did something to destroy the laser coherence. Spinning frosted glass beam interrupters and other techniques are often used to despeckle laser light where it interferes with the laser's intended use.
Speckle does occur with non laser light sources but it isn't usually as strong because of the shorter coherence length. Go outside on a sunny day and look at your fingernails or a piece of black anodized metal, you may see the effect of white light speckle interference. The really amazing part of this technique is that they did it with such a low-res camera. While 42 Megapixels seems like ridiculous overkill for a phone camera, it doesn't hold a candle to the typical spacial resolutions approaching 400 Gigapixels for holography film. Extend and expand the technique into higher resolutions, illuminate with coherent laser light sources and it will be a valuable technique for laser imaging as well as other things. What else has point light sources against a dark sky and obscuring translucent material? The starry night sky. I knew the day would come when finally astronomers can have their cloud filter!