This guy created an Open Source DIY digital camera – including a 1-kilopixel sensor
Dec 31, 2019
John Aldred
John Aldred is a photographer with over 25 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.
Share:
Digital cameras have become so high resolution these days that you might wonder why somebody would want to build one of their own from scratch with a resolution that’s not even a fraction of what digital cameras could do a decade ago. But some things just need to be done because they’re fun.
YouTuber Sean Hodgins has been working on this idea for a long time, and now he’s finally made and released a 1-kilopixel DIY camera. He’s also released the files as Open Source so you can download and make your own, too.
The project is called the digiObscura and was originally designed to be a pinhole camera. After assembling everything together, though, he found that the pinhole didn’t quite let enough light through to the sensors to get a decent reading. So, it was modified to incorporate a dismantled Canon lens.
To form the 1-kilopixel sensor, the camera contains a 32×32 grid of light-sensitive phototransistors. Each of these can spit out an 8-bit value based on the brightness of the light hitting it. A custom PCB contains these 1,024 sensors, which Sean had to solder to the board himself. Although he had the help of a reflow oven to actually solder everything, he still had to place each component by hand, which is a painstaking process.
It’s powered by an Arduino, which doesn’t have anywhere near 1,024 inputs, so in order to support such a high number of sensors, Sean implemented a process called multiplexing. This allows him to read the value of any pixel using only a handful of pins on the Arduino. This is a really slow way of working, but the result is actually quite similar to the way the sensors work in our own cameras, and why we experience things like rolling shutter.
Although the camera produces really tiny images – images that aren’t even big enough to be classed as operating system icons these days – it is a very cool proof of concept device. It’s also a great starting point for anybody who might want to take things a little further.
For example, the camera only captures greyscale images and it does it quite slowly. So, swapping out the Arduino for an ESP32 should allow for a faster acquisition time (thus, lessen the rolling shutter effect). Switching to something like an ESP32 would also allow you to turn it into a WiFi-capable camera that could potentially transmit images as it shoots them. Or you could create your own smartphone app to fire the camera over Bluetooth to eliminate any camera shake from having to manually hit the shutter button on the camera itself – a risk if your camera’s not fixed to something solid.
There are lots of directions you could take something like this. And, sure, it’s not going to keep up with a GFX100, but that’s hardly the point, is it?
All of the PCB designs, the Arduino sketch, 3D printed components and other bits are available to download on the digiObscura GitHub page.
[via Hackaday]
John Aldred
John Aldred is a photographer with over 25 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.
Related Posts
This guy created an open source 3D printable solution to A7III overheating issues
The Focal Camera Is An Awesome Open-Source DIY Modular Camera System
Build an easy DIY overhead camera rig with Open Source Hardware
This DIY USB camera features interchangeable lenses and sensors, and is completely open source
Join the Discussion
DIYP Comment Policy
Be nice, be on-topic, no personal information or flames.
22 responses to “This guy created an Open Source DIY digital camera – including a 1-kilopixel sensor”
ISO?
:)
This should throw a wrench in the sensor size arguments.
But does it have 2 SD Slots???
How’s its eye detection auto focus??? Sony wins again ?
Benjamin Jaworskyj no because that’s the second gen and will cost you double
Why?
Why not?
John Aldred It’s a lot of time, effort and money? To build something that has already been built, is already understood. It’s creating or inventing nothing, progressing nothing in technology. I dunno, it just seems they could have done something better with their time
We all do things that aren’t productive but we personally find fulfilling. I enjoy cooking (seriously, you can just go buy premade food in this day and age) where as experimenting and learning is what does it for Sean, and some seem to find deep fullfillment dumping on how others use their own free time.
Then why do you have a camera at all? Photography is a lot of time, effort and money if you do it right, and you’re just gonna shoot the same images everybody else has. :)
John Aldred I use cameras to create art, or portray my clients products to the best of my ability. I’m trying to create my own perspective, my own view of the world and progress, and improve my skill. To me, doing something like this is like re-making a type writer. I mean, we know they are a thing. They work, there are amazing ones out there, why should I try to make something that has already been made, and much MUCH better than I can make it.
The article is, as far as I can tell, geared at people who think it is a neat project, but lack the time and resources to do it personally. It’s fine that it’s not up your alley, but there’s no need to rain on our parade.
But do you know how they work? For some people, the pursuit of knowledge and understanding is what drives them. Do you not do test shoots and experiment with your gear in order to expand your knowledge and understanding of photography? This guy’s built a camera in order to expand his knowledge and understanding of electronics and how cameras work. Who are you to tell him what his time’s worth or how he should be spending it?
Rich Heaton it’s HIS time and money. Not yours.
(1) If someone isn’t hurting anyone, they don’t typically owe anyone a justification for their choices, but assuming you are just curious and/or puzzled,
(2) For fun. We don’t typically watch tv shows or sporting events to be productive, we do it for fun. Lawyers and doctors don’t typically take photography classes or buy cameras to be productive, they do it for fun.
(3) To learn. Attending school isn’t productive, which is why they had to pass laws requiring children to go to school instead of working. But there is still value (not to mention potential enjoyment and satisfaction) that go along with skill acquisition and general learning.
And a question: if photographic productivity is the only thing that makes anything worth doing, why did you get involved in the comments? These comments aren’t especially educational or particularly fun…
John Aldred I get it, it is his time and his money to do with as he pleases, my thought on the matter is he seems like a super bright guy, why spend all of his talents on the already known? Why not work towards something new or never before done instead of this?
I give up.
Somthing ive wanted to try for a while is making a large format Camera using a flat bed scanner
5.5 axis stabilizing?!
logically, i am expecting this to have crazy iso capability making sony sensor a joke, with such low density and large pixel.
Robert Watson the joke still work on this sensor size i think
I made a DIY digital camera in 1975. I took a 256 X 1 LCD sensor array (state-of-the-art in those days), and coupled it with a fast rotating mirror that would scan the scene, passing it by the linear array. I don’t remember the exact specs, but it took something like 1/30 sec to scan the image, and the whole beast was on a rather large electronic bench, and had to be in a dark room.
I coupled that with a Motorola MC6800 processor to suck the data off, and combine it into an image. All code written in machine code, with a long series of 1’s and 0’s.
It was a fun home project, but I never really imagined the Sony a7R4 that I now use daily…! :)