PDF XChange Forum

I have an image PDF that has barred-i characters (Unicode code point U+0268, also its upper case form U+0197). When I OCR this document, the OCR understandably thinks this is an ordinary 'i'. Is there a way for me to train the OCR system to recognize this character?

This is just one special character out of many that we linguists might want to recognize. So this is not a request to include barred-i in your OCR system, rather it's a question about how I as an end user could add arbitrary roman script characters to the OCR's recognition repertoire, probably for a particular document.

Hello mcswell,

Thanks for your enquiry, however I am afraid that there are no training tools available for our ocr tool

Although this is something we are planning later this year.

Regards,
Stefan

My colleague and myself are both using the same version of xchange viewer for work purposes and yet when I open some documents they font is crisp and clean and hers the font is archaic and jumbles the letters together.
We've gone through all the settings and cant find any differences. Its definitely only happening in xchange viewer as in adobe everything looks as it should.
Help !!

Hi glyon,

I would strongly recommend that you download and try the Editor:
https://www.pdf-xchange.com/PDFXVE6.zip

The Viewer has been replaced by the Editor and is no longer under active development, so only issues that we deem critical would be fixed.

If the Editor displays the same issue, I would recommend that you make sure that the font is embedded within the document, or that the font is installed on her system, as it may be an issue with font substitution because the font isn't embedded in the document.

Cheers,

Was the character training functionality added?

Hello tellabs,

Not yet, but we are in the process of rewriting the OCR tool, so it might get added in the future!

Regards,
Stefan