Newbie and bookscanning software

[Cino]

Hi folks,

I’m quite new in the business and am about to build my own bookscanner. First I checked all the software (OS X) I’d need. Because of the lack of a scanner currently, I started the whole softwareprocess using a pdf file. Finally, using all the appropriate software tools, at the end I’ve got a quite good epub file. My first question is about Scan Tailor. I tried to work with it and noticed that ST is able to split a double page in two. So I wonder why everybody seems to use two cameras instead of one. Why’s this? I mean, the whole construction would be easier and cheaper if I could use just one camera. Any arguments against the use of just one and ST?

And - same topic - does anybody know this guy’s software solution: https://www.youtube.com/watch?v=rjzxlA9 ... ure=fvwrel ? I mean it looks really great, but have no idea what software he uses. I already wrote him an email, but I guess he’s quite the crack and might not have the time to talk to me...

As soon as I start producing my hardware, I’ll document and picture it and will let you know what I do. I hope for lots of hints and look forward to use my own scanner

Cheers, cino

P.s.: please forgive my english... I’m no native speaker!

[markvdb]

Hi folks,

Hi Cino!

I tried to work with it and noticed that ST is able to split a double page in two. So I wonder why everybody seems to use two cameras instead of one. Why’s this? I mean, the whole construction would be easier and cheaper if I could use just one camera. Any arguments against the use of just one and ST?

With only one camera instead of two, you bump into several problems.

Either you scan one page at a time and your resolution stays the same, or you scan two at a time and your resolution halvens. When scanning two pages at a time with one camera second case, you don't want to ruin your book, so you also need to do software correction for the V-shaped cradle you put your book in. This again has a negative effect on image quality.

Quite good results can actually be obtained with very cheap (sub 100€) cameras. See https://picasaweb.google.com/1032615874 ... 1908471714 for example. That's with the standard settings used in our kit, on a prototype that didn't have anti-reflective glass or matte black painted wooden parts.

And - same topic - does anybody know this guy’s software solution: https://www.youtube.com/watch?v=rjzxlA9 ... ure=fvwrel ? I mean it looks really great, but have no idea what software he uses. I already wrote him an email, but I guess he’s quite the crack and might not have the time to talk to me...

Sorry, I don't know about that.

Mark

P.S. One of our customers is documenting his setup with our full kit on Mac OS X...

[victoriaaustralia]

Now I do not know about OsX as I am using a windows machine but I can't speak highly enough of Scantailor and the Homer software program.

I can't understand why more people on this forum aren't raving about Homer - it is so simple and just works every time. Homer can take raw pictures from the camera and use it's own bundled ScanTailor or you can process the pictures in your own stand alone Scantailor and then put them into Homer which uses the Tesseract OCR and pdfbeads to output a very small, OCR'ed file.

I had mucked about with Bindery, ruby environments, PDF beads for Windows, BookScan wizard and all sorts trying to get Pdfbeads to work. HOmer does it all. As I said it bewilders me that more people aren't singing Homers praises. Thanks Heelgrasper.

[daniel_reetz]

Victoriaaustralia, do you have time in your life to do a "short introduction to Homer" to really help everyone understand how great it is? Maybe just some screenshots showing how it works?

(which is to say, I think Homer is relatively unknown here compared to some of the other options)

[victoriaaustralia]

Victoriaaustralia, do you have time in your life to do a "short introduction to Homer" to really help everyone understand how great it is? Maybe just some screenshots showing how it works?

Done! Did it back in March this year:

http://www.diybookscanner.org/forum/vie ... 588#p13588

[victoriaaustralia]

I used Homer yesterday. The input was 812mB of book with text and a mixture of colour and BW plates. Output was a 38mB OCR'd file.

[daniel_reetz]

Pretty awesome. And sorry for forgetting about your previous review!

[JonEP]

Hi - Homer looks promising. Will it work for a two-camera system? Will it work on a 64 bit Windows machine?

Thanks!