I cannot share my document as it contains sensitive data but I made a quick sample to show you my issue.
I'm trying to capture the whole phrase but because there is line going from top to bottom it's like considers the end of the line at the word consider and I'm missing the words "the end of the document". Now if the line was always at the same place I would only capture the first few words or even increase the level of errors allowed. But the issue is that it stops at the line so it doesn't continue after like if the document ended there for the OCR.
Now is there a trick to remove those print lines to improve recognition I already have Thorough pre-recognition and thorough extraction mode. My customer fax in their document so these lines are cause but dirty printer drums or whatever.
Unfortunetaly theres nothing you can do about the lines. But besides increasing error precentage, try adding permit words option in field properties to successfully capture phrases in such cases.
コメント
3件のコメント
Could you please provide a sample image?
Regards
Tim
I'm trying to capture the whole phrase but because there is line going from top to bottom it's like considers the end of the line at the word consider and I'm missing the words "the end of the document". Now if the line was always at the same place I would only capture the first few words or even increase the level of errors allowed. But the issue is that it stops at the line so it doesn't continue after like if the document ended there for the OCR.
Now is there a trick to remove those print lines to improve recognition I already have Thorough pre-recognition and thorough extraction mode. My customer fax in their document so these lines are cause but dirty printer drums or whatever.
Thanks
Unfortunetaly theres nothing you can do about the lines.
But besides increasing error precentage, try adding permit words option in field properties to successfully capture phrases in such cases.
Regards
Timour
サインインしてコメントを残してください。