Hello,
I'm just wondering if this is possible.
We are planning a QF Profile, that OCR's and presents the scan to the operator on screen. The idea is that the user would then "split" the scan into multiple PDF files as required.
What we want to happen next is that the profile "routine" completes a regex scan of the OCR's text, and returns hit to token values. This needs to happen for each individual file.
I'm trying to figure out how this might be acheived as at the moment, I don't know how to "pause" the routine and wait until the operator has split the pages. At the moment the regex ends up with multiple matches and then I have cumalitive token (not what I want).
I hope my question makes sense! Any advice would be much appreciated.
Regards,
Richard.
Question
Question
Quick Fields - Profile that applies regex and token allocation PER document, AFTER scan has been split by user?
Replies
Hi Richard,
So what I'm hearing is that the separation (Spliting) is manual, once split the documents need to go through an automated extraction using regex.
If that is correct, I see this being broken into 2 different possiable configurations:
Ideally you could also incorporate some sort of separation page configuration into the process elminate the need for manual intervention.
1. Quick Fields Only - 2 Sessions
Session 1 - Scan / Import of documents (Twain or Universal Capture) into QF/QF Scanning, ->Manual separation->Store to the repository
Session 2 - Import Session 1 Output (Laserfiche Capture) into Session 2 QF/QF Scanning and run the pattern matching configuration -> User validates extraction -> Store to repository
2. Quick Fields / Workflow
Scan / Import of documents (Twain or Universal Capture) into QF, ->Manual separation->Store to the repository
Then use Workflow to perform extraction using pattern matching ->Route to folder
Hi Troy,
Thank you for your excellent suggestions.
I'm thinking of exploring option 1 - that is having two sessions. Your 2nd option would not work in my sceanrio as what I want is the regex to fill in meta data, then the operator to review/check it is correct & enter any missing meta data before sending it all to the repository. The issue of course is that I want the pages split first, so that means different meta data per document!
I've been doing some reading around using QF scanning and I think I can do my page splitting, then send for reprocessing.
I'll let you know how I get on!
Thanks for your suggestions.