14 replies to this topic

[DSaw]

Larry G posted a raw title

Daubuz, Charles - A symbolical dictionary found @ http://www.biblesupp...-dictionaryzip/


Has anyone given thought to creating a module from this work
or has anyone wanted this as a module

DSaw

[Josh Bond]

If you decide you want to tackle the project, and if you find the OCR text has too many errors (I haven't looked at the text), let me know. I can now OCR a PDF and produce text with far fewer errors than Google Books, while retaining the use of italics (and even bold in many cases)

[DSaw]

HI

I have had omnipage pro for years now I didn't get 18 because because there was no actual improvement to the OCR like 17 had

I love the double click feature for checking that words location in the page of original file from the text editor

I do my own scanning to I have a flatbed so it requires the destruction of the book I cut it apart with a razor blade

It being able to scan PDF's is a big bonus


DSaw

God Bless and Keep you

[Josh Bond]

Good deal, great to have another guy who can digitize books. I have a sheet fed scanner and can knock a book out pretty fast. There's a lot of great material that's never been digitized. I haven't yet mastered removing the spines so i just get a copy shop to cut them off. When I get done with Dr. Peter Pett's commentary, I have quite a few books lined up. Pett's commentary is a lot larger than I thought and quite scholarly--maybe 60+ megs when finished.

[Josh Bond]

We ought to Ocr the same pdf to see what recognition differences there are between 18 and 17. I never used 17 but I'm thrilled with Omnipage 18. It beat abby finereader on every experiment I have run.

[Michael Arnold]

Hey Josh, when you talk about Ocr'ing the same pdf, what exactly do you mean? Please excuse my ignorance.

Edited by Michael Arnold, 15 October 2011 - 11:18 PM.

[Josh Bond]

Software with Optical Character Recognition, or OCR, will create a word processor file (text) from your PDF. Each page of your PDF is just a scan (a digital picture) of a document. The software detects what each character most closely resembles and reconstructs the document into text that you can edit. The accuracy is amazing if the scan is clear.

But don't think you could take the results and turn it into a professor as a term paper. It's far from flawless. Greek and Hebrew characters, when mixed with English, result in gibberish. There will be also be typos in the English, such as "bv" instead of "by". The software algorithms will now reliably exclude headers and footers, along with proper detection of paragraphs and headers, so that the rest of your text doesn't have unnecessary line breaks.

[Michael Arnold]

I have TextBridge. How would I get it to read a PDF file? I thought you had to put a page into a flatbed scanner and scan it using, let's say, TextBridge.

[Josh Bond]

I don't have TextBridge, and have never used it. But generally with OCR software, you can scan directly from a scanner and work from there. Or you can OCR an existing PDF. Since you seem to already have a PDF, just start Nuance. And try dragging and dropping a PDF into Nuance. What are you trying to scan?

[Michael Arnold]

A Greek Grammar, For The Use Of Schools, By E. A. Sophocles, A. M. Revised Edition, Hartford: William J. Hamersley, Publisher. 1867. Digitized to PDF in 2007 by the Internet Archive with funds provided from Microsoft Corporation.