Scribe - extract text from select areas only

[bsharp77]

Hi everyone,

I'm trying to speed up my invoice entry and get as much information in there as I can.

I scan my invoices using CaptureonTouch (Canon's own scanning software) which also does the OCR processing automatically.

I then have a script which imports the PDF to a container and I use Scribe's fileastext command to pull everything out into a field, which is very handy if I ever need to search for any of the invoice contents in the future.

 

What I'd really like to do to take it to the next level is extract text from select areas only and then use the data to populate the fields in the record automatically - such as invoice number, sub total and vat.

I see this can be done using the textract Amazon feature in Scribe OCR, but that adds cost and is very slow in comparison - I don't need OCR as the data is all there anyway, I just need a way of selecting/refining what data it imports.

 

We have around 50 different suppliers and the idea would be to create templates for each supplier's invoice within a script, so that the program will know exactly what data to import on each invoice - like a faux AI system really.

All I would do is enter the supplier code on a new invoice record and the script could then kick in, read the supplier code, match it to the correct coordinates and extract the required data from the PDF in the container field and put them into the correct fields.

 

Is there any way this is possible with Scribe without involving Textract, which is just too slow and cumbersome for me to consider.

Any help or other options much appreciated.