Community

Text figure - many mistakes in OCR

Hi,

I've got source in PDF with numerals as text figure (font: Walbaum). It generates many mistakes (for example often 3 is replaced by 5). I'm wondering is there a possibility to minimalize this mistakes but I'm not sure how to do this.. Should I add some font to Abbyy so it makes it easier to differ numbers?

Greetings Bartek Maciejewski

0

Comments

4 comments

  • Avatar
    Oksana Serdyuk

    Could please specify what ABBYY product you use?

    Also please share your image so that we can test it.

    0
    Comment actions Permalink
  • Avatar
    Bartek Maciejewski

    It's ABBYY FineReader12.

    As example I share image, where numbers 7 in sentence: "Dzieje Żydów w Galicyi i Rzeczypospolitej Krakowskiej 1772–1868," are replaced by 3.

    Example_1

    Greetings Bartek

    0
    Comment actions Permalink
  • Avatar
    Oksana Serdyuk

    Thank you for the provided information. Your question is related to ABBYY desktop product, and this forum works for ABBYY SDK products. I have forwarded your request to my colleagues who are responsible for ABBYY FineReader support. They will contact you very soon.

    0
    Comment actions Permalink
  • Avatar
    Bartek Maciejewski

    Right, sorry for mistake and thank you very much for help.

    0
    Comment actions Permalink

Please sign in to leave a comment.