UiPath Document OCR: Trash Text


I’m using UiPath Document OCR to extract below PDF and I’ve set to apply the OCR. However, the output text is a trash:

Note: I masked some of the information with black crayon.

Below are my OCR settings:

I have asked my other team member to test the document & they got the same trash, even though when we open the PDF, we can select all the texts correctly. However, when I retry digitizing the document after some moment, I cannot replicate this issue again.

I didn’t manage to capture it OCR confidence level. I hope I can avoid this similar issue without retrying the transaction after some time.


Try with different OCR engine and see the result

I can no longer replicate this issue and it works fine now. It happens at an unexpected moment and causes errors to some of the transactions.