PDF Automation that reads handwritten

pdf

#1

Hi guys,

As of now we are developing an automated system on warehousing like
Receiving Orders or Inbound and Outbound Orders, but we are now stuck at some point
RPA can’t read handwritten orders from the receipt scanned.

Is there a way on how to solve this problem using RPA?

Thanks.


#2

I’m pretty sure this is not possible with the current UiPath. This would require externally implemented machine learning algorithms.


#3

Aw so Sad. :frowning:


#4

It’s difficult for even us humans to correctly understand each others handwriting. So you can imagine how hard it would be to build a robot that would get it 100% correct all the time.


#5

Yeah but what if neat and clean handwritten and not like script fonts?


#6

If the handwriting is always in capital letters and consistent I guess it would be manageable with ICR. UiPath has OCR capabilities but that is basically limited to machine typed text recognition from images. Haven’t tried how it reacts to handwriting, probably not too well.


#7

I don’t have ICR on my activities .


#8

My bad, should’ve clarified that UiPath does not support it. ICR is what I meant with external ML.
Don’t know about the future, i’m just a user like you :slight_smile:


#9

Aw can we integrate ICR on UIPATH?


#10

Technically it should be possible (not easy but doable), but from my experience ICR always* requires human verification anyway, so a pure digitizing solution might be better anyway (most of them have ICR interpretation before any human sees them anyway).
Robots are great for accompanying those processes, but personally I view handwritten the same as unstructured input - a struggle between accuracy and %coverage.

While you can get very good results from ICR alone, it’s still risky (“JOHN SMITH” -> “JOHN SHITH”… that customer probably wouldn’t be too happy, and it’s a quite common interpretation error if the M is written with one smooth hand movement).
Biggest risk is always that something is read wrong, but falls within allowed values - there’s no way for a program to know it’s wrong if the value is plausible.

I’d love for someone to pull it off, but let’s remember that even scanned books aren’t recognized 100% (see reCaptcha).

*unless it’s dictionary based data and/or low risk/impact stuff or it’s really good capital handwriting with high DPI scan on an OCR-optimized form… a lot of if’s there unfortunately, while it still can be wrong.

Disclaimer: I may be biased due to previous digitzing experience.


#11

Thanks :slight_smile:


#12

This is not really the domain of RPA … you will probably have to engage with the Machine Learning world … the likes of Abbyy … I think Abbyy are your best bet … amazing company but it’s not going to be super-cheap.