Scanning in your own books

[CharlieBird]

Am I understanding correctly that you guys bought or plan to buy the basic 3600 and not the Plus?
Great thread BTW.

[ascherjim]

Originally Posted by CharlieBird

Am I understanding correctly that you guys bought or plan to buy the basic 3600 and not the Plus?
Great thread BTW.

I know nothing about the Plus. Mine cost about $230. And as I've reported, I'm quite satisfied with it.

[Daithi]

I bought the 3600 and not the 3600 plus. As far as I can tell the difference between the 3600 and the 3600 plus is that the 3600 plus comes with software to make producing PDFs a little easier, but I already have Adobe Acrobat Pro, so I've got that side of things covered already.

[wayrad]

I came to the same conclusion as Daithi and also bought the basic model.

[jayne80]

I use an old flat bed scanner, a heavy object for holding down the book and a copy of omnipagepro 15 or 16 can't remember which, I scan the book straight into omnipage, then block out the text to OCR it and save to .doc format which makes it easier for me to edit.

90% of the time all i need to do to it is check the formatting which in general is just replace the paragraph marks with @@@ then do a find and replace for @@@" .@@@ etc replacing them with paragraphmark" etc then make sure that any chapters are seperated out.

for a 400 page book 4 hours scanning, maybe 4 hours proofing/ocr depending on how many problems i can see, it does all depend on the font of the book as to how well it OCRs. I tend to do the scanning while on the internet or have another instance of Omnipage open to OCR a previous scan.

It may not be perfect but it's good enough to read after being run through mobipocket creator.

[ascherjim]

Originally Posted by jayne80

I use an old flat bed scanner, a heavy object for holding down the book and a copy of omnipagepro 15 or 16 can't remember which, I scan the book straight into omnipage, then block out the text to OCR it and save to .doc format which makes it easier for me to edit.

90% of the time all i need to do to it is check the formatting which in general is just replace the paragraph marks with @@@ then do a find and replace for @@@" .@@@ etc replacing them with paragraphmark" etc then make sure that any chapters are seperated out....

It may not be perfect but it's good enough to read after being run through mobipocket creator.

I have come to the (perhaps obvious) conclusion that whichever version of FineReader one uses, it's better to do one's editing with an external word processor rather than attempting to do it within the FineReader software (which I believe anyway is only possible -- but unwieldly -- with the Pro 10). While I have both Microsoft Word and Corel WordPerfect on my computer, I have found WordPerfect in fact to be far easier to effect the final formatting. And for some reason only the basic Sprint, and not apparently the heavily-featured Pro 10, permits transferring the scanned images to WordPerfect.

I am puzzled, though, by your description of replacing "the paragraph marks with @@@ then do a find and replace for @@@" .@@@ etc replacing them with paragraphmark." What does this accomplish?

Anyway, to emphasize, I have found formatting my titles with WordPerfect much faster (and easier) than with Word. But maybe it's just because I'm possibly more adept at using WordPerfect than Word. We all have (evolved) our separate methodologies for this book scanning and formatting procedure.

[Daithi]

My OpticBook 3600 scanner arrived today. I've only got to play with it a little bit, and I'll be at my girlfriend's this weekend, so I won't get to play with it in earnest until Sunday afternoon. However, my initial impression is pretty good. I scanned in a picture of an old soda fountain image and it looked outstanding. The same image scanned with my old cheap scanner had looked absolutely awful. I can't wait to really start playing with this thing.

I did have one problem though and thought I'd ask if anyone else has encountered this. I opened up Photoshop and selected Import and there were two options, one is for the OpticBook 3600 and the other is for a WIA compatible device. The Opticbook 3600 selection does the preview, provides lots of control over the scanning process (I just selected the defaults), but it returns an error that there is not enough RAM to complete the scanning. The WIA option scans just fine but without the control of the OpticBook 3600 option. My laptop is fairly new and has 3GB of memory, so this doesn't make a lot of sense. Has anyone else had this problem?

[charleski]

Originally Posted by gazza

So my idea is to scan in a book a day and proof-read -- a pleasant hobby --

Sorry, but I had a good chuckle at this. The first book will be an amusing diversion. The second book will be a technical exercise. The third book will be tedious. The rest will be drudgery.

There's a reason people get paid for proof-reading, believe me. Obtaining the raw text is quick and easy. Transforming that into something truly readable that is similar to a professionally-produced book is a lot of work. Do not under-estimate the amount of effort involved here.

Originally Posted by Daithi

In my case, I'm scanning old books that I don't want to destroy. This means cutting off the binding and using a flatbed scanner is out of the question. I don't even want to press real hard on the book to get it to lie flat, because I'm afraid I will break the binding and have my pages falling out.

Do what Google does. Use a camera with copystand lights instead of a scanner. Lay the book on one of those book stands that allows the book to open at 90 degrees and position the camera perpendicular to the page. If you just have 1 camera, photograph all the even pages first, then reposition it and do all the odd ones. You may find it helps to get a piece of glass to place over each page to flatten it (google uses a fancy infra-red auxiliary scanner to detect page geometry without flattening pages). Here's a pic of Google's scanning method. You can do something similar without all the fancy extras they use. Since all you're doing is turning pages inbetween each capture this method places a lot less strain on a book's binding than flatbed techniques.

[alecE]

Oooh! What a wonderful thread! I've been playing with my old flatbed scanner and Abbyy Scan to Office for some old Penguin books of mine. the results have not been great so far because of the distortion at the spine. (And I just cannot bring myself to destruct the books). Despite owning a Canon camera, tripod, tether, usb cable, macro flash, remote release, macro lens, I seem to have omitted purchasing the nugget of common sense which says "Use the camera, dumbo!"
Smacks head, thanks to everybody who stated the obvious, smacks head again.
(Do I get any karma for providing a snigger of amusement to everybody else?)

[igorsk]

Try the latest FineReader, it has special preprocessing for straightening the lines.

[HauntedAttic]

I've been using OmniPage 15 for about a week. I found it slightly more accurate than Finereader Sprint 6, and the automated workflows are great once you've set them up. I also found that that at 400 dpi it will correctly identify em dashes (at 600 dpi it simply skips them).

For my purposes, its tendency to combine one-line paragraphs with the following paragraph is simply unacceptable. It can throw off the author's intended meaning and is hard to spot. Imagine the confusion that results when the dialog from one character is combined with the reply from the next.

The latest version of OmniPage may have this issue solved, but without a trial version I'll never know.

Right now I'm testing FineReader 10 Pro and it seems to do everything I need. Like OmniPage, it can automate tasks, strip headers and footers, and save to RTF without excessive formatting. I also like being able to proof the text while viewing the image. Unlike OmniPage 15, it does all this without annoying quirks and crashes.

Looks like a winner. Now I just need to grit my teeth and break out my credit card.

[ascherjim]

Originally Posted by HauntedAttic

...Right now I'm testing FineReader 10 Pro and it seems to do everything I need. Like OmniPage, it can automate tasks, strip headers and footers, and save to RTF without excessive formatting. I also like being able to proof the text while viewing the image. Unlike OmniPage 15, it does all this without annoying quirks and crashes.

Looks like a winner. Now I just need to grit my teeth and break out my credit card.

I've just finished my testing of FineReader 10 Pro, comparing it with Sprint 6, and have let the test version lapse. I found that attempting to proof the text while viewing the image within the Pro 10 actually more unwieldy than proofing within the word processor I sent the scanned text to. (As I reported earlier, I prefer WordPerfect over Word, although I have both.) Also, surprisingly, Sprint 6 seems to have more dpi options than the Pro 10. And the Sprint 6 also has the option to send the scanned text directly to WordPerfect, which, again surprisingly, the Pro 10 doesn't. On the Sprint 6 I can generally eliminate headers and footers through its blocking option. The bottom line for me is that the Sprint 6 is working just fine, if not better than the Pro 10, but if anyway the two were equal for me in performance, the cost of the upgrade would decide me against it. Good luck.

[HauntedAttic]

I agree, upgrading to FineReader 10 is a luxury. I did my first book with Sprint 6 and a flatbed scanner. The results were so good that it convinced me I could convert more of my old paperbacks to RTF.

Right now I run the same TIFF through FineReader 10 and OmniPage 15 (came free with my scanner). Then, after using VBA to clean up and standardize the output, I use Word to compare the documents. It's the quickest way I've found to proof the book, although the process still takes hours. The two programs both make mistakes, but they rarely make the same mistake on the same character.

I try not to think about the fact that I probably could have done the same thing with Sprint 6 and saved myself the price of the upgrade.

[ascherjim]

Originally Posted by HauntedAttic

I agree, upgrading to FineReader 10 is a luxury. I did my first book with Sprint 6 and a flatbed scanner. The results were so good that it convinced me I could convert more of my old paperbacks to RTF.

Right now I run the same TIFF through FineReader 10 and OmniPage 15 (came free with my scanner). Then, after using VBA to clean up and standardize the output, I use Word to compare the documents. It's the quickest way I've found to proof the book, although the process still takes hours. The two programs both make mistakes, but they rarely make the same mistake on the same character.

I try not to think about the fact that I probably could have done the same thing with Sprint 6 and saved myself the price of the upgrade.

I've just received from ABBYY a Christmas(?) offer of the Finereader Pro 10 upgrade at half price! For $89. Even though I've been using my Sprint 6 quite happily, I couldn't resist this offer. The email offer to me is:

http://webmail.aol.com/30044-144/aol...?ws_popup=true

If that doesn't work, let me know.

[ascherjim]

Originally Posted by ascherjim

I've just received from ABBYY a Christmas(?) offer of the Finereader Pro 10 upgrade at half price! For $89. Even though I've been using my Sprint 6 quite happily, I couldn't resist this offer. The email offer to me is:

http://webmail.aol.com/30044-144/aol...?ws_popup=true

If that doesn't work, let me know.

I suspect that link won't work: Try the following, either to purchase through download or by box.

http://infonews.abbyyusa.com/redir3/...ROMO=HOLIDAY09

http://infonews.abbyyusa.com/redir3/...ROMO=HOLIDAY09