Daniel Reetz, the founder of the DIY Book Scanner community, has recently started making videos of prototyping and shop tips. If you are tinkering with a book scanner (or any other project) in your home shop, these tips will come in handy. https://www.youtube.com/channel/UCn0gq8 ... g_8K1nfInQ

Laser-cut stand to digitize newspapers

Built a scanner? Started to build a scanner? Record your progress here. Doesn't need to be a whole scanner - triggers and other parts are fine. Commercial scanners are fine too.
Post Reply
User avatar
right2read
Posts: 6
Joined: 19 Apr 2013, 15:32
Number of books owned: 0
Country: Argentina
Location: Buenos Aires
Contact:

Laser-cut stand to digitize newspapers

Post by right2read » 27 Mar 2021, 00:22

Hi! In bibliohack.org we are working on a stand to digitize newspapers. The need to have very low-priced equipment to digitize newspapers, which in many cases are in danger of being lost (especially in small cities, I'm talking about Argentina) is a very persistent demand. We are working on a model made of 9 and 6.5 mm MDF, for laser cutting (very easy to find suppliers here).

Image

Here is the project developed in freecad:
https://github.com/Bibliohack/Cenital

Here a 3d View
http://files.bibliohack.org/models/ceni ... 310b.xhtml

And here is a video where you can see how it will work (and other fun things)
https://twitter.com/BiblioHack/status/1 ... 3707022340

cday
Posts: 291
Joined: 19 Mar 2013, 14:55
Number of books owned: 0
Country: UK

Re: laser-cut stand to digitize newspapers

Post by cday » 27 Mar 2021, 09:11

Congratulations on your novel design optimised for a specific use!

It would be interesting if you could provide a short overview of the main features of the design: deducing details from images is sometimes not too easy, especially when they change while being viewed.

Imaging a large page satisfactorily requires more MP to maintain sufficient DPI, are you planning on using a camera with a particular MP specification?

Is it planned to slide the newspaper sideways to image left and right hand pages, or do you plan to image all left hand pages and then all right hand pages, and then interleave the images?

The lighting system is interesting, do you intend to use LED or fluorescent tubes? The idea illumination for imaging is parallel light as from the sun, achievable in principle with a point light source and a parabolic reflector, multiple parallel lamps is a more practical potential alternative approach to providing even illumination which should result in a more compact design than the lamps used on most other scanner designs.

Is it intended that the glass (or acrylic?) sheet used to flatten the newspaper pages be raised and lowered by hand or by a motor?

What is the purpose of the double-hinged plates with the circular holes each side?

It will be interesting when the hardware is built and refined to see a typical page image, I expect that you are looking forward to that too!

dpc
Posts: 340
Joined: 01 Apr 2011, 18:05
Number of books owned: 0
Location: Issaquah, WA

Re: Laser-cut stand to digitize newspapers

Post by dpc » 27 Mar 2021, 19:36

Seems overly complicated for what you need to do. You basically need a copy stand with some way to hold the pages flat. In the simplest case, the camera, and even the lights, could be positioned on tripods instead of the complex system you've shown. That being said, overkill doesn't necessarily hurt you though. If you're set on building something that elaborate to photograph newspaper pages it may very well do the job.

My biggest concern however is the use of a single camera to capture the entire newspaper page. I would think that you'd want a minimum of 300 DPI for archival purposes. Broadsheet paper is 23.5"x29.5" in size. At 300 DPI that's (7050px x 8850px) 62.4 megapixels, and that assumes perfect 100% framing of the shot. I suspect you'll find that to get a camera of that high resolution, you could purchase a large format scanner outright that would do what you want. So what this means is that if you still want to use a single camera (~20MP) for this scanner and get 300 DPI resolution you'll need multiple smaller-sized shots per page and then stitch the resulting photos together in a post-processing operation.

This is where you can spend more time with your design to allow the positioning of the page under the camera to make repeatable indexed moves to capture smaller portions of the page. A word of advice - before you do any sort of scanner design, you should get out a camera and a tripod and take some test shots of a sheet of newspaper and then try to see what software is out there that can stitch things back together properly. I wouldn't even bother with hanging lights as you can just photograph things outside and use sunlight and pin/tape the newspaper to a large piece of cardboard lying on the ground to keep it somewhat flattened. Once you figure that post-processing out, begin work on the design for your moveable indexed base. You could build a prototype of that piece of your scanner and take that outside as you did the previous experiment and get the kinks out of the indexed movement before moving on to the easier parts of the scanner (camera & lighting mounts, hinged platen glass, etc.).

User avatar
right2read
Posts: 6
Joined: 19 Apr 2013, 15:32
Number of books owned: 0
Country: Argentina
Location: Buenos Aires
Contact:

Re: laser-cut stand to digitize newspapers

Post by right2read » 28 Mar 2021, 00:54

Congratulations on your novel design optimised for a specific use!
thanks!
It would be interesting if you could provide a short overview of the main features of the design: deducing details from images is sometimes not too easy, especially when they change while being viewed.
yeah, sorry! now I don't have enough time, but in a few days I can post more details with more images, but in the twitter video you can see how the digitization process is
Imaging a large page satisfactorily requires more MP to maintain sufficient DPI, are you planning on using a camera with a particular MP specification?
We have to use a Canon Rebel with 24MP, this gives us a capture area of 51x34 cm (20' x 13,3') with 300ppi, that is ok for part of the material to be digitized (one shot per page, the base can slide and move the newspaper around the digitization area), some newspapers that are larger will be digitized between 250 and 300 dpi (1 star FADGi performance level allows up to 150 dpi!, but 230dpi would cover the entire possible area according to the dimensions of our device with this camera, the idea is that the base of the device is interchangeable and in the future we will design others according to the needs that arise, but the area cannot overcome 65x100 cm)
Is it planned to slide the newspaper sideways to image left and right hand pages, or do you plan to image all left hand pages and then all right hand pages, and then interleave the images?
The camera support can rotate to allow both options, but if we need to use the total area, it will be to digitize posters no newspapers, and in that case it is not important to have a high resolution, but rather an uniform lighting (I think that in this case we are going to have a problem, we'll see)
The lighting system is interesting, do you intend to use LED or fluorescent tubes? The idea illumination for imaging is parallel light as from the sun, achievable in principle with a point light source and a parabolic reflector, multiple parallel lamps is a more practical potential alternative approach to providing even illumination which should result in a more compact design than the lamps used on most other scanner designs.
initially we will use cheap led lights like this, then trial and error until we get a satisfactory capture, if necessary we will place better lighting sources (according to the time and money available for the project)
Is it intended that the glass (or acrylic?) sheet used to flatten the newspaper pages be raised and lowered by hand or by a motor?
i dont know, It can be either one (glass or acrylic), I assume we will use the cheapest
What is the purpose of the double-hinged plates with the circular holes each side?
yeah, sorry for this, it is a sarrus link with hinges, its seem too much, but it is a way that makes the light source always aligned, and that factor does not depend on the operator's attention, I do not know if it will work, if it does not work the lights will be fixed in the corners ¯\_(ツ)_/¯
It will be interesting when the hardware is built and refined to see a typical page image, I expect that you are looking forward to that too!
we really have to build something this month or else we'll have problems, this project was funded last year but was suspended due to the pandemic (we are a year late!)

User avatar
right2read
Posts: 6
Joined: 19 Apr 2013, 15:32
Number of books owned: 0
Country: Argentina
Location: Buenos Aires
Contact:

Re: Laser-cut stand to digitize newspapers

Post by right2read » 28 Mar 2021, 01:43

Seems overly complicated for what you need to do. You basically need a copy stand with some way to hold the pages flat. In the simplest case, the camera, and even the lights, could be positioned on tripods instead of the complex system you've shown. That being said, overkill doesn't necessarily hurt you though. If you're set on building something that elaborate to photograph newspaper pages it may very well do the job.
Yes! I think exactly the same! but it is the product of various learnings using this shitty material (lasercut mdf), learning about how operators use the devices, and learnings about how you can have a stand built from a distance in cities where certain materials are more difficult to find (lasercut mdf is cheap and easy to get suppliers)

but with lasercut mdf it is difficult to build copy stand like this!, we build this:

Image

(more info here, in spanish)

Although it was functional and we digitized some things, it had many problems (instability, difficulty in accommodating the lights, difficulty in accommodating the camera, deformation of the material, etc.)

we made other models, but smaller like this

Image

(the device doesn't have the lights or camera in there, but this little copystand actually digitized a huge amount of museum archive material using just a powershot and chdk!)

and this

Image

(a copy of this was cutting and assembled in another country, by people without any technical background, who were literally learning that it was a screw while they were assembling it, and after that we learned that if you can put the pieces together like a puzzle without tools, then really anyone can build it! more info here)

So, we wanted to get out of the stereotype of the copy-stand that we conceive and try something else, maybe it works better with this material, but we can only know if it works when digitizing in the context of an institution and with real operators
My biggest concern however is the use of a single camera to capture the entire newspaper page. I would think that you'd want a minimum of 300 DPI for archival purposes. Broadsheet paper is 23.5"x29.5" in size. At 300 DPI that's (7050px x 8850px) 62.4 megapixels, and that assumes perfect 100% framing of the shot. I suspect you'll find that to get a camera of that high resolution, you could purchase a large format scanner outright that would do what you want. So what this means is that if you still want to use a single camera (~20MP) for this scanner and get 300 DPI resolution you'll need multiple smaller-sized shots per page and then stitch the resulting photos together in a post-processing operation.

This is where you can spend more time with your design to allow the positioning of the page under the camera to make repeatable indexed moves to capture smaller portions of the page. A word of advice - before you do any sort of scanner design, you should get out a camera and a tripod and take some test shots of a sheet of newspaper and then try to see what software is out there that can stitch things back together properly. I wouldn't even bother with hanging lights as you can just photograph things outside and use sunlight and pin/tape the newspaper to a large piece of cardboard lying on the ground to keep it somewhat flattened. Once you figure that post-processing out, begin work on the design for your moveable indexed base. You could build a prototype of that piece of your scanner and take that outside as you did the previous experiment and get the kinks out of the indexed movement before moving on to the easier parts of the scanner (camera & lighting mounts, hinged platen glass, etc.).
in the previous post I answer some of these questions, it is possible that later we will make tests of digitizing maps (up to 90x60 cm) in two parts by moving the base and then using a software that joins the images, but it is not a priority for now

Our development cycle (to call it somehow) involves building prototypes, testing them and making changes to them (and we make a lot of changes and we test a lot of ideas as you can see), seeking to arrive at a stable model for a certain type of use

(So far we have only achieved a stable model in all these years, but it is v-shape, and cheap copystand is a permanent demand from institutions here)

User avatar
right2read
Posts: 6
Joined: 19 Apr 2013, 15:32
Number of books owned: 0
Country: Argentina
Location: Buenos Aires
Contact:

Re: Laser-cut stand to digitize newspapers

Post by right2read » 28 Mar 2021, 01:51

So, we wanted to get out of the stereotype of the copy-stand that we conceive and try something else, maybe it works better with this material, but we can only know if it works when digitizing in the context of an institution and with real operators
Needless to say, it doesn't matter if the apparatus seems complex, if it can be assembled like a puzzle without using tools, and if it can be cut at any cheap laser cutting provider, then it will still be possible to replicate the assembly by anyone without technical knowledge.

dpc
Posts: 340
Joined: 01 Apr 2011, 18:05
Number of books owned: 0
Location: Issaquah, WA

Re: Laser-cut stand to digitize newspapers

Post by dpc » 28 Mar 2021, 20:06

Please post an update once you've completed a scanner and worked through all of the kinks. It's helpful for others that visit the site and are looking to solve the same sort of issues (newspapers/posters/large page sizes). Thanks!

Post Reply