A new scanner design using plastic tubing

[dpc]

Do you see any advantage in having the cameras mounted as part of a separate structure?

Yes, In my case there are advantages to not moving the cameras. My cameras are relatively heavy compared to the point-n-shoot compacts used by most in these forums. I have a pair of Canon DSLRs that I put on tripods on either side of the scanner. Doing this allows me to reposition the cameras depending on the page size (my scanner supports up to 12" x 14" pages) to utilize more of the available camera sensor resolution. I also don't like the idea of my cameras moving while I scan books. I will typically use longer shutter times to shoot with a reduced ISO in an attempt to reduce graininess in the images so the less chance of some settling after turning a page and lowering the platen the better. I also don't have to worry about cables attached to the camera (USB/power) flexing each time I move the platen.

There shouldn't be a focus problem between the front and the back of the book (I haven't seen one, anyway and the math can prove it). I don't scan particularly thick books, so I made a mock up this afternoon using a cardboard box underneath a large gardening book that would emulate a 4.25" thick book and used an 8.5x11" page with text from my laser printer as the page content for scanning.

Below are subrects of the images of what would be the first page of the book vs. the last page of the book. This is the untouched jpeg image coming straight from the camera and I don't see any difference in focus between the two. Measuring from the page surface to the image plane of the camera was the same distance (27.75") in both cases. The only thing that moved between the shots is the cradle moving horizontally so that the platen would come down into the book's gutter properly from first page to last page (as most all scanners do).

First page

first_page.jpg (93.61 KiB) Viewed 8942 times

Last page

last_page.jpg (106.26 KiB) Viewed 8942 times

[davidlandin]

This next weekend I am going to try a couple of Nikon DSLR cameras on the scanner. As you say, they are heavier than the point and shoot variety of camera, but I think by adjusting the counterbalance weights there should be no problems. We'll see what happens!

David Landin

[bsbob]

Interesting about the 90 degree bend. Separate sounds like the way to go then.


Did you have to wire those lights in yourself or do they plug into a standard outlet?


Is there such a thing as too much light? I'm thinking more lights... More light, brighter pages, less shadows... Maybe more chance for reflections though.

[Mohib]

I've just bought a low cost LED floodlight. It cost about $20 US on eBay

I've ordered the 10w version. What are you finding the colour temperature like on yours (I presume you ordered the cool-white?). Could you post some samples using the best white-balance setting for your camera for your light? I've searched the archives here and these have come up now and then but I've really wanted to see what kind of results (colourwise) these lamps give (as best as that is possible to show given variations in white-balance calibrations from camera to camera). Also I've heard their temperature changes a bit as they heat up or after being on a while (how long a while is I don't know). Have you noticed such drifts in colour?

Thanks.

[davidlandin]

I've just bought a low cost LED floodlight. It cost about $20 US on eBay

I've ordered the 10w version. What are you finding the colour temperature like on yours (I presume you ordered the cool-white?). Could you post some samples using the best white-balance setting for your camera for your light? I've searched the archives here and these have come up now and then but I've really wanted to see what kind of results (colourwise) these lamps give (as best as that is possible to show given variations in white-balance calibrations from camera to camera). Also I've heard their temperature changes a bit as they heat up or after being on a while (how long a while is I don't know). Have you noticed such drifts in colour?

Thanks.

I'm at an early stage with trying out this new light Mohib. And I'm going to be away for a couple of weeks. So I will try and report back when I'm home again and have had a chance to do some more experimentation. However I will say the light looks promising. Please let me know how yours works.

David Landin

[davidlandin]

Did you have to wire those lights in yourself or do they plug into a standard outlet?
Is there such a thing as too much light? I'm thinking more lights... More light, brighter pages, less shadows... Maybe more chance for reflections though.

The lights came with a length of three core electric wire. I added a plug, and plugged it into a wall socket. Pronto! Lights on!

You are right - good lighting is essential to avoid shadows etc. But another factor is to do with optics. I think with lighting the principle is this. . . With brighter lights you can have a smaller camera aperture - a smaller hole to let the same quantity of light into the camera. A smaller aperture gives the camera greater depth of field. This means that things are sharply in focus lets say from 30cm - 40cm (just imaginary figures). If you have less light, then you need a bigger aperture to let the same amount of light into the camera, and a bigger aperture means less depth of field in sharp focus. So this might be from only 35-36 cm from the camera. When you are imaging text, you need to have a crisp, sharp focus on the text, and this is easier to achieve if everything from 5cm above the page to 5 cm below the page is going to be in focus. Focussing is much more difficult to achieve with a shallower depth of field, where a movement of just a centimeter difference could result in an out of focus image. Not explained that very well, but maybe it helps.

David Landin

[rkomar]

Cheap point and shoot cameras (e.g. Canon A series) are fixed aperture, so there's really no way to change the depth of field on them. You just have to be very careful about setting them up perpendicularly to the platen if you want good focus across the book page. A brighter light does help with keeping noise rates down on those same cameras, which is a good thing.

Here's another tip I learned the hard way: if you use a hot light like a halogen bulb, keep it from shining on the camera and heating it up, as more dead pixels appear the hotter the sensor gets.

[daniel_reetz]

Cheap point and shoot cameras (e.g. Canon A series) are fixed aperture, so there's really no way to change the depth of field on them. You just have to be very careful about setting them up perpendicularly to the platen if you want good focus across the book page. A brighter light does help with keeping noise rates down on those same cameras, which is a good thing.

yup, that's right. you want the lights bright enough that you can use ISO 100 (or some other suitably low sensitivity) without the neutral density filter dropping into place.

[Ralf]

David, thank you very much for your great design!
After many years of considering which scanner to build, I will finally try to build one!

As I couldn't find the right plastic tubes here at home markets in Austria, I will try to make it out of wood (got a friend who is carpenter, so probably it's even cheaper).

Before I start I hope you can help me with the corrrect measurements. You uploaded two documents, "Scanner diagram.pdf" and "Scanner diagram 11 x 14.pdf", in both it says "The sizes on this structure are made to fit a plastic platen with each side of the platen measuring 22 cm x 35 cm".
22 cm would be 8,7 inches, 11 x 14 inches would be 28 x 35 cm...

So is the platen in the "bigger" model 28 x 35 cm?

Thanks a lot,
Ralf

[bsbob]

Yes, some kind of larger/adult tinker toy wooden structure could work. Funny. I've thought that a few times when looking at the pvc pipes.

Then again, the cost of pvc pipe is pretty cheap.