Invoice Process Automation

Hi, I’m new to UiPath. i have to create a invoice automation which means extract a data from different types of invoice image pdf file and put it into a SAP.Please tell some open source OCR for effective recognition of image type pdf and give the efficient method and steps to create an invoice automation. Thanks in Advance. @arivu96 @aksh1yadav @badita @ovi @Vlad

Try with Abby and I am not sure about the accuracy of extracted data.

1 Like

In my opinion, Abby Cloud OCR is the best engine you could use. But you need a paid license to run it.

Regards,

1 Like

thanks for your reply. Pls suggest some open source OCR which is best with UiPath. @Lucas.Pimenta

whichone to i install abbyy finereader or flexi capture. @Lucas.Pimenta @harinathreddy.yn

I haven’t used Abby yet bro so I am not able to suggest you, will see if any one who has used it may suggest.

Open source OCR’s will not work as expected.

1 Like

Thanks for your reply Bro…

Hi Prakash, did you try the Abbyy Flexicapture already? How was it?

In case you have a variety of formats, I think you should also consider using OCR APIs, there is a couple of AI-based services available in the cloud that have the advantage that they do not have to be manually configured and are template-agnostic.

There is a couple of such cloud services available - like Abbyy Flexicapture for Invoices (a bit different from normal Flexicapture), SMACC or infrrd.ai. But I think the best option to try (in terms of accuracy, traction and ease of setup) is Rossum Elis, which integrates with UIPath directly and quite easily - see for an example https://rossum.ai/blog/2018/07/30/automating-data-extraction-from-invoices-using-rossum-api-and-uipath/ .

(Disclaimer: I’m affiliated with Rossum.)

Hello!

I tried using this and it works fine, except i need to also use details regarding Product Description and I cannot find an appropriate Extracted field type for the same. Could you suggest me as to how I can possibly do that .

Thank You,
Anusha

Hello Anusha, the product description is often part of the line items, which Elis already attempts to discover (though the accuracy is lower than for header fields yet) so check that out. Another option is to use Rossum’s verification interface and process that field semi-manually.

Cant we customize those fields?

Also I tried your Rossum Verification Interface. I got all the required values in the table segment,Is there any way to access the table values in API so that i can extract all my data for automation?

Hi Anusha, how does the Elis OCR work? Is it working fine for you? Thanks

Hello @kenzaburo,

I came through certain disadvantages, so i switched to Abbyy Flexicapture.