Forgot your password?
typodupeerror
Technology

Gigapixel Camera Catches the Small Details 61

Posted by Soulskill
from the i'll-order-a-dozen dept.
An anonymous reader sends this quote from Nature: "David Brady, an engineer at Duke University in Durham, North Carolina, and his colleagues are developing the AWARE-2 camera with funding from the United States Defence Advanced Research Projects Agency (abstract). The camera's earliest use will probably be in automated military surveillance systems, but its creators hope eventually to make the technology available to researchers, media companies and consumers. ... AWARE-2 sidesteps the size issue by using 98 microcameras, each with a 14-megapixel sensor, grouped around a shared spherical lens. Together, they take in a field of view 120 degrees wide and 50 degrees tall. With all the packaging, data-processing electronics and cooling systems, the entire camera is about 0.75 by 0.75 by 0.5 metres in volume. The current version of the camera can take images of about one gigapixel; by adding more microcameras, the researchers expect eventually to reach about 50 gigapixels. Each microcamera runs autofocus and exposure algorithms independently, so that every part of the image — near or far, bright or dark — is visible in the final result. Image processing is used to stitch together the 98 sub-images into a single large one at the rate of three frames per minute."
This discussion has been archived. No new comments can be posted.

Gigapixel Camera Catches the Small Details

Comments Filter:
  • celebrate! (Score:5, Funny)

    by Sparticus789 (2625955) on Wednesday June 20, 2012 @05:39PM (#40391323) Journal

    I've always wanted an in-flight UAV to be able to diagnose me with skin cancer.

  • .....Roseanne Barr beaver shot at an exquisite level of detail is technically possible!

    • .....Roseanne Barr beaver shot at an exquisite level of detail is technically possible!

      Just because a thing is possible doesn't mean it should be done. Also, eeeeeeeeeeewww.....

    • .....Roseanne Barr beaver shot at an exquisite level of detail is technically possible!

      According to Larry Sanger [slashdot.org], Wikipedia already has that image.

    • There are some things money can't (or shouldn't, or should be illegal, or should get your house firebombed for suggesting) buy. For everything else, there's couchslug.

    • .....Roseanne Barr beaver shot at an exquisite level of detail is technically possible!

      According to Larry Sanger [slashdot.org], Wikipedia already has that image.

  • by mpeskett (1221084) on Wednesday June 20, 2012 @05:40PM (#40391339)
    Coming soon to Minecraft?
  • Great... (Score:5, Funny)

    by jmerlin (1010641) on Wednesday June 20, 2012 @05:41PM (#40391347)
    Just when I thought I had plenty of space for my HD porn.
  • by Anonymous Coward

    Lets enhance that! [youtube.com]

    • by PPH (736903)

      There! That's the one that did it! I'd recognize that pixel anywhere!

  • by Anonymous Coward on Wednesday June 20, 2012 @05:43PM (#40391373)

    Can't they just stack a shitload of gpus and parallelize the stitching? Where's the bottleneck?

    • by timeOday (582209) on Wednesday June 20, 2012 @10:14PM (#40393639)
      RTA, they're working on a 10 hz version now. Although I think "stitching" is a bit misleading, since it normally implies compute-intensive pattern matching to register the images over each other. In this case it seems like you'd only do that once, since the lenses are fixed, and then just re-use the same mapping.

      Anyways, that works out to about 18 MB per sensor per frame, times 100 (sensors) times 10 (hz) so 18 GB / s. So I think they'll need hardware video compression as well to do much with it. Otherwise you're filling a 4TB hard drive in a little over 3 minutes.

      • by timeOday (582209)
        PS, obviously a hard drive cannot write that fast. Assuming 60 MB/s minimum sustained write speed [macperformanceguide.com], you'd need to stripe across 300 drives to push 18 GB/s :)
        • Just to point out your errors: Your own link shows the deskstar having a 85MB/s minimum sustained write speed, and if you were to design a system that a minimum speed was required, you could quite easilyuse 2 drives, with one writing from inner track to outer, and the other from outer track to inner in a RAID-0, and as your graph shows, it would sustain 400MB/s between the two. You could also leave the inner tracks unused. Or use a faster drive that spins at 15k RPM. You still aren't near the speed you w

          • by timeOday (582209)
            Balancing writes to inner and outer tracks does seem like a clever idea - I wonder if somebody does it. But it would not be trivial to do correctly. RAID0 continuously writes an equal number of blocks to each drive, so you get twice the speed of the slower drive, so it would be as if you were writing to the slow inner sectors all the time if you did it with RAID0.

            Also, the throughput writing files to a filesystem is somewhat slower than writing blocks to a disk. There is some bookkeeping overhead, such

    • Can't they just stack a shitload of gpus and parallelize the stitching? Where's the bottleneck?

      Well considering 3 frames is about 3 GB of data, I would guess it takes that long to write the data to a harddisk. Accordingly, to the article they are still working on filtering out the data they do/don't want.

  • by Anonymous Coward

    Busch Pressman Model D with a Nikkor lens, provia 100f, and a 4800ppi scanner. Works like a charm. The 6GB TIFF files are a bit unweildy though.

    • by graphius (907855)

      so you have a very detailed image of film grain?

      • by Anonymous Coward

        not really, he's describing a medium-to-large format camera. even 35mm scans fairly well up to 1GP, with the right stock.

        but really, you'd want something like the Hulchers. they take 70mm roll film and burn through an amount best measured in feet per second, each image between roughly 6x9 and 6x17 cm. of course, that's pretty much what they flew in the U-2 and friends: it's not the camera, it's the platform -- and every pound carried takes away from flight time.

        this is all about reproducing that level of de

        • by mug funky (910186)

          the very best 35mm is still below 1GP. look at the MTFs. just because the grains are smaller, doesn't mean the dye clouds left behind after processing are, or that the light hasn't diffracted to all hell on its perilous journey through the film.

          the sharpest stocks are the thinnest, and thus have bugger-all dynamic range.

          the ones with the best dynamic range tend to be the fastest stocks, which are grainy and not too sharp.

          digital has beaten film for quite a while now. it's just a matter of making it work as

          • Actually with t-grain films (Like Delta 100), sharpness and good dynamic range are available in the same stock.. Digital has beaten 35mm consumer film, equalled 35mm or 645 pro film (at a higher price), and still does not come close to large format. It is however superior in convenience and productivity and delivers good enough results for a fraction of the effort, for most needs.

        • Actually, Busch Pressman model D is a 4"x5" camera... which is 10 x 13 cm.

    • by mug funky (910186)

      you'd be hard pressed to have as deep a field as this camera though.

      or get as good a dynamic range on such a high contrast stock.

      or have as good MTF characteristics (3-layer film is actually piss-poor at this, as much as i love the look of it).

      • by toadlife (301863)

        The GPs post and your reply are reminiscent of every vinyl vs CD debate/flame war I've ever seen.

        In ten years, when we all have gigapixel cameras in our phones, are we going to have film devotees going on about how uniform pixels cannot possibly convey an image in the same way that film grains can?

        • They've already been doing that for years. So far, they have been proven right. Maybe in ten years the price of top of the range sensors has dropped sufficiently for the outcome to be favorable for the sensor camp. Right now, you need prohibitively expensive sensors and large format cameras to come up with a picture that is better in image detail, color trueness and resolution than you can get with the same quality mechanics/optics and a film at a smaller size than the sensor you're competing with. Until th

          • by Mal-2 (675116)

            With digital, you can take a hundred pictures for the cost of the charge in a set of batteries. To do the same with 35mm, you'll need roughly three rolls, plus processing, and you have to change film, are stuck with the type and speed of film in use, and can't see your results until they come back from the lab. Thus, learning to take good pictures with a digital camera is far cheaper and faster than learning to do it with film. Also, very few people will ever see your film directly. Either it must be printe

  • by JustOK (667959)

    Blink and you've still got a few seconds before you miss it.

  • by J-1000 (869558) on Wednesday June 20, 2012 @05:59PM (#40391543)
    I know this is inevitable and a nifty advancement, but it creeps me out. It's bad enough to have low-resolution cameras that can point in your general direction and maybe zoom in if they want to be extra nosy; now they'll have ultra high-resolution cameras with a 120 degree field of view that can identify you two blocks away.
    • I know this is inevitable and a nifty advancement, but it creeps me out. It's bad enough to have low-resolution cameras that can point in your general direction and maybe zoom in if they want to be extra nosy; now they'll have ultra high-resolution cameras with a 120 degree field of view that can identify you two blocks away.

      They already have cameras that can identify which brand of cigarette you're smoking from 80 miles away. Look up sometime. The problem isn't that the government has this technology, the problem is what they do with it.

    • I can *already* identify you from two blocks away with a prosumer grade camera and a crap kit lens.

  • by dogsbreath (730413) on Wednesday June 20, 2012 @07:04PM (#40392029)

    a camera that truly matches the detail zooming capability of webcams on NCIS, CSI whatever and so on.

  • Instead of camera array? The latter is more flexible in terms of being able to have a controlled tradeoff between resolution and other parameters as needed, such as extended depth of field, capturing depth information, extended dynamic range, and others related to lightfield photography and computational photography.
  • Translation. (Score:2, Insightful)

    by Alex Belits (437) *

    Each microcamera runs autofocus and exposure algorithms independently

    Translation: we just duct-taped together a bunch of standard camera modules, and did not bother figuring out how they work.

  • The top of the article says it's equivalent to 1000 cameras, but we've had 10MP cameras for close to a decade now. Even the article itself (farther down) says that it uses 98 (a hell of a lot more than 1000) 14MP camera sensors.
    • Your math, flawed it is.

      Also, it's not the megapixels that count, but the field of view and the details captured. Most of that is in the lens and the combining of the sensors.

      • Your math, flawed it is.

        Care to point out the flaw?

        Also, it's not the megapixels that count, but the field of view and the details captured. Most of that is in the lens and the combining of the sensors.

        Why is it the "field of view" that counts? Your required field of view is entirely dependent on A) The size of your target and B) The distance you are from your target.

  • Zoom in. Enhance. Zoom in. Enhance. Zoom in, enhance.
    Gotcha!

    And for some reason they'll want a print for good measure; print out !

  • How can all the 96 micro sensors share the common spherical lens, and still have independent autofocus independently?
  • Isn't this the Lytro camera [lytro.com] concept?
  • by Anonymous Coward

    DHS AUTO-TRACKING OF US CIVILIANS TO BE CITY WIDE

    Watch ---

    Wide Area Airborn Surveillance: Opportunities and Challenges - Gerard Medioni

    on youtube. What about showing the auto-tracking software used to track all moving
    objects across the entire frame or the video of automatic object
    recognition tracking every car on the freeway, or perhaps all this is
    why all the license plates are blurred and there are no aerial
    perspectives? DHS appears to be attempting to integrate WASS (wide area
    surveillance system) and Pers

Arithmetic is being able to count up to twenty without taking off your shoes. -- Mickey Mouse

Working...