I was thinking about this yesterday while looking for resources on building a board to read data from a modern sensor from something like an a7 (turns out its way over my head). The issue you’d run into with the pi hq camera module assuming you can get the focal plane right and everything to fit is that the sensor is way smaller than 35 film so you’ll have a rather large crop factor.
I just really wish Raspberry Pi would come out with a truly HQ camera with a full frame 35mm sensor, or at least an APS-C sensor. I'd pay $1000-1500 for it, just to have something as good as a full frame camera but programmable and hackable.
I would love to build a full frame mirrorless camera that runs my own UI. I'm pretty sure I could code a much more advanced UI than Sony or Canon.
Their current HQ camera is more like an LQ camera and there is not a huge variety of high quality photographic lenses available for it.
I would love to have a nice APS-C or full frame board-level camera that can be easily integrated into something like a Raspberry Pi. At present it is nearly impossible to get something like that --- dev kits for large sensors from Sony Semicon, Canon, and ams are profoundly expensive, only sold to companies, and go through a lengthy quoting process.
> The sensor alone doesn't deliver 100% of the result of a modern SLR. The image processing code does a lot of heavy lifting that is mostly proprietary.
This is HN, hackers like us love to be able to tinker with the heavy lifting. That's the whole point of the Raspberry Pi cameras. Even with the current HQ camera, a used point-and-shoot camera with a similarly-sized sensor (or even bigger sensor) would tend to produce better images and videos while being more compact, robust, convenient, and cheaper than the DIY camera.
I would be very surprised if the camera modules put out anything different than the standard image sensor interfaces such as SPI, MIPI, and SLVS. Anything else would require extra circuitry that is too big. Then it just becomes a case of figuring out the control for the lens module.
APS-C might be a tough ask, but something like this[1] or this[2] might work - smaller sensor but at least you won’t get a ridiculous crop factor on APS-C lenses. The second one says it’s format is APS-like.
But I’d say an RPi alone would be a bit tough to handle something like this. You’d need at least an FPGA to control the sensor and maybe something like RPi for UI and further processing.
But it’ll be expensive. These components are anything between $500 and $3000, and that’s _without_ the FPGA and other things that would be required. An equivalent crop frame camera might be only $1000(?). If the aim is to have a DIY hackable camera, sure. If you want a cheap camera, no...
Everything you described can be built out with this existing sensor and hardware.
You don't need an APS-C or larger sensor to get decent images. Most APS-C sensors use a different high-speed interface that won't work with the Raspberry Pi anyway.
Really, this solution from the Raspberry Pi foundation is a great start for any of the projects you mentioned. It's also cheap and highly available.
This is great. I don't care at all about the 16 megapixels (to misquote Bill Gates, 6.4 megapixels should be enough for anyone), but AF and reasonable close focusing have been available in cheap smart phones but missing from the Pi til now. There hasn't even been a usable way to read a QR code with a Pi until now. The $25 price is great too.
The Pi foundation should have shipped something like this long ago. Maybe they can work out something with Arducam to make it a standard Pi product.
You can also interestingly get the raw sensor data from the Pi camera too, as in before the demosaicing is applied.
I wonder, this may be silly, but could you not potentially get a slightly higher resolution image, if you just created a pixel for every sensel. Than if you were to apply the demosaicing algorithm then convert to grayscale.
I wonder whether this technique could be adapted to produce a something like a digital equivalent of the Fuji/Hasselblad XPan camera, or a 617 camera (with an aspect ratio of about 3:1).
The stumbling block with conventional sensors is that no one wants to commit to a production run of an odd shape of sensor.
Compared to smartphones, outdoor cameras need to be built to very different specifications and standards. I don't think my iPhone would last very long duct-taped to the side of my house.
Of course, neither would a camera that's made out of cardboard and runs on a Raspberry Pi. But for prototyping, this seems like it could offer a good start.
I've daydreamt before of a device, sensor attached by thin film ribbon cable to a faux film canister holding the computery bits. A general fit solution to film camera conversion. But, sensor thickness and power source may be intractable problems.
Thanks for the link to the Kodak sensor - that's awesome, because those sensors have a fairly common interface across all the other Kodak sensors like the 48x36mm Medium format 22MP sensors... It would be pretty easy to adapt this project to make your own CCD medium format camera!
I've often daydreamed about a sub-unit that could fit in the back of a film camera and contain a sensor and some electronics. The sensor part would be thin enough to fit over the pressure plate and the electronics would be in the film can space along with a micro sd card for storage. The existing aperture/shutter would continue to control the light reaching the sensor. Basically digital film. That would be neat.
reply