How to extract circled boolean fields in Document Understanding


Hello everyone,

Whats the best method I can extract the attribute above in intelligent form extractor? Is there another extractor i can use to extract this field?

That is interesting.

If you can catch that element under Tessaract OCR detection, very less chance though. you might need to create a custom model and train for that.

Below video would help:

Building & Training Custom ML Models for Document Processing | RPA | UiPath - YouTube

Also, just a thought, if you are using an OCR extracts, the extracted text will have some character before and after “No” and if yes is circle there will be some character before and after “Yes”

So you can create a logic according to that, may be?

Hi Rahul

Thank you for your reply. I did apply a signature logic to the boolean field (assuming the circle was a signature) but the extracting confidence is low. The only way seems around it is through an ML extractor and UiPath has a documentation to it too.

However, I only have one field to extract and wondering if there is any other way to extract other than deploying an ML skill and custom model.

Give a try to find that character thing can be useful?

It it’s only one field then going for Document Understanding path is not a good idea.