There was an earlier thread on this scanner which I have not revived at
https://www.mobileread.com/forums/sho...highlight=czur
The Czur Scanner is a budget priced book scanner which was funded on Indiegogo. Whilst usually too conservative to risk my money on a crowd funded project, I was impressed with the promotion and had been so frustrated at previous attempts to scan books that I made an exception. I have now received my scanner and am very pleased with it.
I'm not going to go into enormous detail. The scanner has a number of different modes. The only one I will mention here is via connection with a computer running custom software downloaded from the Net on a suitable version of Windows. I believe a Mac version of the software is in development.
Basically the scanner looks very like a lamp, with a camera at the end of an arm. A black mat comes with the scanner and is placed adjacent to the scanner under the arm with the camera. On selecting scan on the software and selecting book mode, the black mat will be displayed on the screen and the lcd. In other words, a completely black background. The book is placed face-up and open on the black mat. It can be held open and as flat as possible using one or both thumbs in the middle of the left and/or right edges, but not at the top or other locations. It seems that the software allows for this. A press on the foot pedal and the page is scanned. You can then turn to the next page and repeat the process. This can be done fairly quickly, though the promotional video is of course probably an exreme. The software identifes the outline of the book on the black pad, and in book mode compensates for the presecnce of the thumb or thumbs, the curvature of the pages etc. Each image will of course include two book pages. I elected to have the software split these two pages, a simple process. I then loaded these split pages and ran the OCR software to produce a Word document. The results were excellent for a hardback (out of print and no ebook) and a Paperback (also not in print). I then edited the document in Textmaker, which has an excellent export to epub function. A quick spell-check caught most errors, which were surprisingly few. Some errors not detected by the spell-check were obvious on a quick read and fixed via search and replace. For exemple, on one book the software seemed to mistake "the" for "die" on occasions. I also took the trouble to divide the book into chapters and set suitable headings, which enabled Textmaker to generate a TOC. It is early days yet, but it seems to me that it should be possible to produce a vary usable epub consisting only of the text in a very short amount ot time, by simply scanning and spell-checking. Of course, more time would be required for a more polished result, including perhaps things like a cover, Chapter Headings, Footnotes, a TOC etc, and of course correction of any more subtle errors.
I am surprised and pleased with the results. I now have ebooks of two out of print books where no ebook has been made available. These are of course for my own private use and for the sake of my poor eyesight. The hardback book in particular has such small print that I would be very uncomfortable trying to read it.
I partly repeated the process for two pages of one book and have posted some images. One image is the original scan of 2 pages. Two images are those two pages as split by the software. The quality of the images has of course been reduced to comply with Mobileread guidelines. Finally I have attached the word document produced by the OCR software run on the two split pages.
It's early days so far, but I'm very pleased with my purchase at the moment. Not perfect, but a huge improvement over anything else I've tried. It seems that it may actually make scanning some of those books where there is no ebook a viable process. And spare my poor eyes in the process.