DIY Single Pixel Digital Camera

[Artlav] wanted to build a digital camera, but CCDs are expensive and don’t respond well to all wavelengths of light. No problem, then, because with a photodiode, a few stepper motors, the obligatory Arduino, and a cardboard box, it’s pretty easy to make one from scratch.

The camera’s design is based on a camera obscura – a big box with a pinhole in one side. This is all a camera really needs as far as optics go, but then there’s the issue of digitizing the faint image projected onto the rear of the camera. That’s fine, just build a cartesian robot inside the box and throw a photodiode in there.

There are a few considerations when choosing a photodiode for a digital camera. Larger photodieodes have higher noise but lower resolution. [Artlav] has been experimenting with a few diodes, but his options are limited by export control restrictions.

Even with the right photodiode, amplifying the tiny amount of current – picoamps in some cases – is hard. The circuit is extremely sensitive to EMI, and it’s inside a box with stepper motors pulled from the scrap bin. It’s amazing this thing works at all.

Still, [Artlav] was able to get some very high resolution images across a huge range of wavelengths. He’s even getting a few images in mid-wave infrared, turning this homebrew digital camera into the slowest thermal imaging camera we’ve ever seen.

46 thoughts on “DIY Single Pixel Digital Camera

  1. From whatever limited resources (I mean I don’t know, bare CCD sensors from China these days ‘are’ pretty cheap, I think on almost anyone’s budget), but starting on a similar project myself, I do think this is pretty cool. And where my own aims are also towards the ‘digital + retro’, good to know we are not alone in saying: “Take That ! Instagram…”

    1. Gotta hand it to the Guy for using what he has on hand for sure… Pretty clever actually, if a bit rough in some aspects. Gotta give him props though for being determined, and actually getting some really awesome results.

  2. He could clean the images up a bit by putting a straight line inside the camera obscura. When included in the image recoding process, one can correct misalignments in-software by using this line as a reference.

  3. This is really cool. It allows him to tailor the spectral response of the camera as needed, simply by changing the detector. Variations on this imaging technique can (and has) been done at gamma-ray, X-ray, UV, LWIR, THz, and mmWave wavelengths, although the camera obscura setup won’t be appropriate over all of this wavelength range.

  4. “but his options are limited by export control restrictions” … I was laughing so hard! Must be Google Translate … He’s in Moscow, with pretty much unlimited access to anything in terms of electronics, inlcuding locally manufactured – he’s descriping an IR photodiode he got, manufactured in SPb, Russia. He was also talking about getting a proper material lens, and ended up getting ZnSe from what sounds like eBay – maybe that’s what Google translated as export control restrictions, haha.

    And yes, his 3D printer is something!

    1. Aliexpress and Chinese ebay have lots of stuff from cheap PCs and 5$ phone cases to turbomolecular pumps and lathes.

      Too bad the author thinks that if Obama says sanctions forbid me to buy stuff from the US, then I can’t buy it somewhere else.

    2. Actually the “export control” part was about Japan – Hamamatsu refused to sell me any of their IR sensors without some serious paperwork, since they can apparently be used for making homing missile guidance systems.

      1. Okay, I made a stupid ass comment earlier, but i don’t think you will be able to buy that one anyway, it usually requires a request from an established company with a traceable history and a paperwork trail.

        Try looking for Flir sensors, I think they can be bought off ebay easily, but I am not sure if it is tge same wavelength you need.

    3. FYI There are things you cannot get from the US in western europe due to export restrictions, not as civilian.

      And I’m not talking about guns, or drugs, I mean electronic devices.

  5. I’m guessing from the movement of the clouds that he’s scanning different wavelengths separately, changing out the detector each time. If he were to put multiple detectors on there next to each other, he could scan them all more or less simultaneously (he’d just have to fix the offsets.)

    I love the look of the photos, they have a nostalgic, surreal quality to them.

  6. I love this – quite the thought provoking project. Just out of curiosity, would there be any advantage to moving the very simple pinhole “lens” and keeping the complex sensor still? I know, light leaks etc. but still, moving a flat bit of board is pretty simple.

      1. thanks for correcting, i crashed on that line for a while pondering if it could be correct.
        great find too, I’d love to see more of this, sadly the build details are a little skimpy.

  7. I built an AA (ascii art) version of this waaaay back in the day, except I used an CdS LDR, “didgitised” it with a 555.
    exactly the same idea, except I used a 6″ PVC end cap, a short piece of 6″ pipe with a vertical slit, then a piece of dog food can with holes 45 degrees apart rotated by a stepper, the LDR in the centre of the can.
    That took care of the vertical axis, then swept the assembly left/right with another stepper.
    The whole thing was controlled by my SYM-1 and images printed out on an Epson MX-80.
    Took ages to get it working, the steppers were 7 degrees/step, even with half stepping the resolution was pretty low.
    BUT it did work!
    I got the idea from watching the camera obscura “effect” that my bedroom window curtains created on the opposite wall.

    I’m going to have to try this with an Arduino and output to a Raspberry Pi

    1. Without clever rigging, it would add the complexity of coordinate transformation. – You still want the final image to be displayed in XY.
      You could apply a barrel distortion to the existing image to clear up the fringes. Compared to this, a spherical scanning rig would do the scan a little faster.

  8. Please try to copy the technical information correct. Lately you mess up the small amounts of information you provide quite often.

    “Larger photodieodes have higher noise but lower resolution” -HackaDay

    “The smaller the photodiode is, the higher the resolution is.
    The bigger the photodiode is, the lower the noise is.
    It’s your choice.” – Original

  9. Instead of a cartesian scanner, he should employ a swing that scans the sensor along a cylinder. It would fix the perspective errors at least on the horizontal plane. If he could make the sensor move along a spherical surface, it would look even more natural.

    Since we’re moving the sensor anyhow, one might put a long black straw in the hole with the other opening on the sensor, and put the pinhole on the end of the straw so the hole always stays parallel to the sensor. That avoids the problem of the hole changing its shape when seen from different sensor positions.

  10. I’d steal another idea from Newtek and put in a color wheel on a stepper motor, so it could automatically capture color (actually, as many different bases as you had filters…)

  11. awesome

    really, by the description of the camera I didn’t expect much, but the IR pictures are very interesting, potentially even useful and some of the pictures are quite charming and offer a surprising level of detail. Well done!

Leave a Reply to Thinkerer Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.