Invoice Processing for Noisy documents

I want to use intelligent ocr/ machine learning for scanned pdfs… I have tried scrapping using OCR… It is not giving good results… Any suggestions on how to use Intelligent OCR/ML for multiple pdfs?

Any video available?


You can try tesserract, which is an activty, or click here. for other ocr intelligent activities or here for examples directly.

please come back to us and let us know if this works.

Happy automation,

don’t forget to like the post and mark it as solved if this answer solved your problem


1 Like

I tried using Intelligent OCR activities…But the document I am using doesnot have lines between columns of the table… Probably because of this, it is unable to get the required fields… What can be done in this?

Can I train multiple invoices with same keywords?