When labelling an invoice, the selector includes the line below it. The value is then made up of different characters, which is wrong. (HABNu1n6 6t0e2 2r01H5522O)
Because its select the next row I cannot get those values out either.
Any suggestions on how to fix it?
Can you tell me which activity you are using ?
I working on a labeling documents in document understanding.
We are using the Australian invoice model as a base
Are you able to drag a rectangle around the title in stead of clicking it?
Perhaps this way you can only select the correct line.
Regards
Soren
This might be an OCR issue, is the text in the PDF readable or does it need to do OCR to extract?
What does the raw text look like if you can check that? You need to use one of the PDF activities to check.
When I drag it still selects the bottom line.
Looks like the PDF is readable. Does that mean we dont use DU to extract data from these ones? How much code will you need to build to do that?
Hi @jacko.smit
I think you can extract data using simple activities like Read PDF Text. The code needed is minimal - just steps to read the text and extract required fields using regex or string methods. DU is more useful for scanned or complex PDFs.
Happy Automation
Does the extracted text show the header and the values below formatted in any weird way?