remove Noise from Text

Dear support,

i try to use FRENGINE ABBYY OCR in the recognition of the text (french langage) The result is not bad but sometimes we get a noise in the text  words that contains special characters (like "_éé_$".I want to know how  i can  improve the recognition (cleaning the text) without using text mining or AI.




1 comment

  • Avatar
    Denis Gusak


    The first way to do it is you can add words with special characters to the dictionary. Thus the word that is in the dictionary will be recognized more accurately. Please check [Help → Guided Tour → Advanced Techniques → Working with Dictionaries]

    The second way is you can create you user pattern. Please check [Help → Guided Tour → Advanced Techniques → Using GUI Elements → Recognizing with Training].



    Comment actions Permalink

Please sign in to leave a comment.