Convert with OCR number only scanned pages

I hav a lot of pages scanned in from old statistics to convert to digital format.

I have used gocr but the conversion is so full of errors (confusion with alphabetic characters) that i wonder if nobody ever wrote a digital characters only OCR program.

Anybody a hint?

on the amiga there was an OCR in fxScan as far as I remember.

Other than that, I recommend Readirs
http://www.irislink.com/mail/1309/RI14/Mac/LP-RI-1309-us.html

there is a Windows and macOS version. The linked version above is offered at a special price ATM.

@JosDuchIt

Depends on what format you need it in you could use one of the many free online converters, like this one

It does .doc, .txt, .pdf and .rtf

I'm sure there are others who can do more...saying that you can always run the resulting file through another converter that does exactly what you need (as long as it's nothing amiga specific, like amiga database files and such)

JosDuchIt : could you share a sample with one or two pages in private way ? I could try other soft solution at my work ...

@_bigdan_
Thanks i posted links with PM

@amogo1 & Raziel thanks for suggestions

I just dicovered gcor has an option for onverting only to numbers
-C "0123456789"
There are other options that may be usefull to finetune the conversion
Studying their effect

@JosDuchIt

You have PM

@JosDuchIt


you have a second PM..

all the pages are scanned since a few days.

The pages you provided are scanned with low resolution, the results are not so optimal, but if you can re-scan with higher resolution, I can process them with character recognition software again.

@Amigo1

Thanks,
I did send you a private mail twice, did you receive them?

@Amigo1
Quote:

I haven't used it for a while, but yes, there is an OCR function in fxScan and it is said to be adaptive.
This means that initially you may get "read errors" which you have to set aright manually. After a while of usage, the OCR function in fxScan produces less "read errors". IIRC, best scanning resolution for OCR-scans is 300 dpi.

Just for the records - fxScan also has a "create PDF" function, but I'm not sure if it is still usable today.
The last PDFs I created with it were not readable on e.g. WIN-PCs - I guess the PDF format fxScan uses is outdated meanwhile.
Can someone confirm this?

@JosDuchIt

Hi I did log into this site today. I did find and sent you the TXT files. let me know if you got them!

@Dandy

I bought FxScan 5.x years ago. What puzzles me, I had 2 CDs but I can't find any of them anymore.

It might be it was using PDF v1.2 or v1.4?

@Amigo1
Quote:

Bummer!
But 2 CDs for FxScan 5.x?
Did you have two copies of it, or were the two CDs 'FxScan 5.x' plus 'Betascan' or 'ScanQuix 5'?
As fxSCAN required an free 'Betascan'-driver or an already installed 'ScanQuix'...

But 'FxScan 5.x'?
From my memory I only know fxSCAN versions up to v4.05 - will have to check when I'm back home tonight...

Quote:

Hummm - I created and edited a 'test.pdf', but cannot see any obvious errors:



The PDF-version generated by fxSCAN seems to be v1.2.