Most performing OCR program

Convert page images into searchable text. Talk about software, techniques, and new developments here.

Moderator: peterZ

Post Reply
georges44
Posts: 2
Joined: 31 Aug 2016, 04:15
E-book readers owned: kindle
Number of books owned: 1000
Country: france

Most performing OCR program

Post by georges44 »

Hi everybody,

I have been lurking at this forum for many years before registering.

Up to now, i have used a flatbed scanner to digitize about 15 books, and then used either scantailor or OMNIPAGE OCR to assemble the PDF.

My goal is to switch to a custom build book scanner which will allow me to greatly increase my workflow, without, if possible, compromising on the quality.

My attitude is the results of the DIY book scanner should be as close as possible to the quality obtained through a flatbed scanner, in order to perform an OCR, and compress the book to a maximum.

I have compared the performance of the various paid/free OCR programs.

In my experience, Omnipage OCR has consistenly given the best OCR results.
The programs i tested were tesseract (i have used the windows gui sunnypage ocr), acrobat pro, abby finereader, and pdf x-change editor.

This doesn't seem to have been adressed on these boards, and thus i'm curious to know if experienced members here have come to the same conclusion as me.
BruceG
Posts: 99
Joined: 14 May 2014, 23:17
Number of books owned: 500
Country: Australia

Re: Most performing OCR program

Post by BruceG »

I also use Omnipage. It came with a scanner a few years ago so have kept using it upgrading as needed. Have tried Acrobat v9 but had no way of editing.
Have used Omnipage mostly for books, minutes and magazines to produce pdf files. Magazines and books with the same subject I index with Acrobat to make searching quick and easy. Have done the same with minute books. Currently having a go at creating epub for a ereader.

A set of Newsletters from the 30's and 40's have been the most difficult. They were produced with a Roneo machine. It may have been quicker to just type them out again. I was glad when the typerwriter was replaced or the Roneo machine.
georges44
Posts: 2
Joined: 31 Aug 2016, 04:15
E-book readers owned: kindle
Number of books owned: 1000
Country: france

Re: Most performing OCR program

Post by georges44 »

I agree; attempting to OCR a bad quality printed text is a nightmare. Did you try increasing the resolution?
BruceG
Posts: 99
Joined: 14 May 2014, 23:17
Number of books owned: 500
Country: Australia

Re: Most performing OCR program

Post by BruceG »

I expect I scanned at 300dpi as usual. It was the first and only time I had scanned typed and duplicated pages. The ones kept may have been the worst ie. the good ones were distributed. I used a typewriter font to look more like the original.
L.Willms
Posts: 134
Joined: 21 Sep 2016, 10:51
E-book readers owned: Tolino Shine
Country: Germany
Location: Frankfurt/Main, Germany

Re: Most performing OCR program

Post by L.Willms »

I am using ABBYY Fine Reader 11 and I'm satisfied.
thickglass
Posts: 1
Joined: 18 Dec 2016, 09:22
E-book readers owned: PocketBook( all kinds of)
Number of books owned: 100
Country: USA

Re: Most performing OCR program

Post by thickglass »

Omnipage works the best for me, ABBYY Fine Reader 11 is a bit worse
vergnuegt
Posts: 3
Joined: 12 Dec 2016, 12:39
Number of books owned: 500
Country: Germany

Re: Most performing OCR program

Post by vergnuegt »

In the moment I use Finereader 11 and 12 and used Finereader Sprint 6.0 and 9.0. I am satisfied.

One year ago, I tried Omnipage SE. It´s a smaller package of the full version. The results are o. K., but I don´t like that Omnipage SE installs an updater in the Windows control panel.

I think, both programs do the job right, but I only have experience with the german language modules.

vergnuegt
Post Reply