Linear Book Scanner – Open-source automatic book scanner (2014)

Hehe nice. There is a whole community about this topic at: <a href="https:&#x2F;&#x2F;diybookscanner.org&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;diybookscanner.org&#x2F;</a><p>Years ago I once wrote a little tool in Java called bookbuilder, where you could turn the pages manually, make a photo and then run an automatic process on all images to build a searchable pdf.<p>I used <a href="https:&#x2F;&#x2F;boofcv.org&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;boofcv.org&#x2F;</a>, an impressive Computer Vision library in pure Java, still exists and it is pretty fast, too.<p>It was able to detect the page contour, deskew it, flatten the image and remove finger contours by matching the skin tone, then build a PDF with integrated invisible OCR Layer without any user interaction. I remember that I was working on line slope detection with some kind of watershed algorithm to improve the flattening part.<p>Fun project, I wonder if I have the source code laying around somewhere... even the download page is gone today. This was long before I went open source with all of my little side projects, because I never thought it could be interesting for someone else :-)

There‘s a somewhat similar commercial scanner [1] [2], with a V-design as well but inverted to scan from the top. Much gentler on the books as it‘s the scanner that moves, not the books themselves. Super happy to see someone develop an open-source alternative!<p>[1] <a href="https:&#x2F;&#x2F;www.treventus.com" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.treventus.com</a> [2] <a href="https:&#x2F;&#x2F;youtu.be&#x2F;SdipuAuWsEs?si=dFWRtva5gO2oM91o" rel="nofollow noreferrer">https:&#x2F;&#x2F;youtu.be&#x2F;SdipuAuWsEs?si=dFWRtva5gO2oM91o</a>

I love the idea, although the risk of torn pages is mildly concerning for archival purposes or valuable books. Though if that were the case, I&#x27;m sure scanning by hand would be preferred anyway. I&#x27;ve often wanted a device like this for the purpose of digitizing my excessively large collection of books.<p>Regarding frequency of torn pages in the FAQ:<p>&gt; Prototype 1 could scan the majority of books without damage, but may tear one or two pages in some books. Out of 50 books tested, 45% had one or two of their pages either torn or folded. This is a very early prototype and there are many areas for improvement in the design.<p>In my opinion, this is mostly acceptable. Especially if a future revision reduces the 45% to somewhere around the ~10-20% range. If I had the space for a device like this, I would definitely consider building one.

This is a problem domain where software hasn&#x27;t caught up with what is possible, so people do in hardware what could be done in software.<p>With two or more photos or a stereo image (new iPhone?) one could triangulate to infer a flattened page, and produce images that look like they came from cut pages in a flatbed scanner. Now just pay someone well in Ethiopia to carefully turn pages without damage.<p>As any researcher can attest, our digital libraries now hold a century of scanned work of questionable quality. AI could infer scans indistinguishable from an outline font format original on an 8K monitor.<p>I once helped consult on the 1980&#x27;s font wars, turning old formats and digital scans into Postscript and TrueType fonts. This was hard then, but will soon be understood as the &quot;correct&quot; way to scan text, when software catches up.<p>For the scientific literature, we need a ChatGPT equivalent to reconstruct LaTeX source that can reproduce each page. (We really need a successor to LaTeX that isn&#x27;t such an arcane language, and can author fixed and flowable text with equal ease.)

This looks a lot safer<p><a href="https:&#x2F;&#x2F;www.inforum.com&#x2F;newsmd&#x2F;ndsu-students-book-scanner-inspires-spinoffs-around-world" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.inforum.com&#x2F;newsmd&#x2F;ndsu-students-book-scanner-in...</a><p><a href="http:&#x2F;&#x2F;diybookscanner.org" rel="nofollow noreferrer">http:&#x2F;&#x2F;diybookscanner.org</a>

Just curious - Once we scan, we have all contents in digitized format. So, why unbinding a book to pages before scanning is not a scalable model? Is this to avoid additional work of unbinding?

This seems like it’ll shred any book that’s even slightly damaged..

Google already did all the books, to a first approximation. Problem was copyright and owners of it.<p>They used low paid labor to flip the pages. I&#x27;ll send a link later if no one beats me to it.<p>Back home: it&#x27;s <a href="https:&#x2F;&#x2F;www.theatlantic.com&#x2F;technology&#x2F;archive&#x2F;2017&#x2F;04&#x2F;the-tragedy-of-google-books&#x2F;523320&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.theatlantic.com&#x2F;technology&#x2F;archive&#x2F;2017&#x2F;04&#x2F;the-t...</a>

This looks absolutely fantastic. Not only can you digitize your books, but it also shreds them for you for free!<p>Pretty sweet 2 for 1 deal.

I am allergic to dust mites. After about six months on a shelf, I can&#x27;t touch a book without getting an allergic reaction. Simply dusting or even vacuuming the exterior doesn&#x27;t work. This gizmo looks like something I can use to rid a book of dust mites so I can read it.

Of course this would be possible with DRMd books. One could, hypothetically of course, use e.g. some kind of script to automatically turn the page of the e-reader, screenshot it, then use image recognition to convert to epub, etc. to archive them.

This is cool, easily scan my books at home that i will use to train my own LLM! Super me!

I tried it but not found it useful.

Fantastic concept! I agree with the concerns about damaged pages. Perhaps this is something that could be easily improved.

tl;dr a diy system for scanning books. Basically, build a triangular prism with a special zig zag slit for a single page to snake through. Put a vacuum on each side to pull the paper through the slit. Put two optical scanners in the middle of the slit, to scan both sides of the sheet as it hangs down. Attach a motor to move a sled pushes just the top, then bottom, of the book. The site features six designs. Some include videos of operation [0].<p>[0] <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=84byulcC6i4">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=84byulcC6i4</a> 30 seconds long<p>In the fullness of time, maybe I would make one of these, given that I live in an apartment and not a house with space to construct&#x2F;store this scanner. I check etsy every couple of years and haven&#x27;t seen someone offer a kit. I use 1dollarscan, though they&#x27;ve had to restrict their offering as Pearson, et al notice their existence.

just saw the binding off and push it through a scansnap or the like