To make sharpen better of scanned pdf

Hello, everyone!
I’m working on a project which is get numbers from a pdf file.
Using Read PDF with OCR and tesseract(digits.traineddata) but can not get the exact number.
Can’t understand “8” or “9” from the pdf.
Even tried export pdf as image and crop image and used tesseract ocr but still now working.
Do you have any ideas?

Regards,
Temuulen

Hi @Temuulen_Buyangerel ,
Have you tried UiPath document OCR?
image

regards,
LNV

It says:
UiPath Document OCR: Request CorrelationId: eb033a5a-264a-4082-8b5f-8f451526f3a1
Request PredictionId:
Error performing OCR: Invalid API key specified UiPathOCRInvalidApiKey
I need API key right?

Yes, but in read PDF OCR
it will gen API for you

I just tried but still doesn’t have api key on it.

Hi @Temuulen_Buyangerel

Use Read PDF with OCR activity and change the scaling of tesseract OCR engine. Intially it will be blank start from 0 and increase by 0.5 and see how the text is getting extracted. Keep the profile in scan.

Hope you understand!!

1 Like

I took api key from the document understanding, and it works fine.
Let me check all cases.
Thank you!

Hope it working for your file
cheer
LNV

1 Like

OMG it works 100% on all of my files.
Thank you!!
cheer

Congrats @Temuulen_Buyangerel ,
Thanks for mark my way to solution
regards,
LNV

1 Like

UiPath document OCR got some limits of using? like 50 calls per hour
regards,
tmk

In the document understanding, I do not see. but reading a file i think this is not the problem
I never see limit with it, hope it has no limit :smile:

1 Like

Oh, okay.
Thank youu!

1 Like

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.