You are viewing limited content. For full access, please sign in.

Discussion

Discussion

Quick Fields 9 issues

Quick Fields
Updated May 28, 2015
posted on April 30, 2015

I have a couple of weird things happening with Quick Fields 9 (It does have the latest patch installed).

1. I have quite a few classifications with Zone OCR. I use Laserfiche Capture Engine. I can bring in a sample page, program my pattern matching and it brings in the account number and invoice number without any issue. When I go to process that same exact page that I used as a sample, one or both of the account numbers/invoice numbers will either be blank or will have the %(invoice) in the field. Or it will have some gibberish text in the field.

2. Every so often when a document doesn't classify and it is a 2 page document, I will split the documents i Quick Fields. Sometimes the "split" option is greyed out and it won't let me split. I then have to import into Laserfiche, split the document and then reprocess with Quick Fields. Any insight into this will be helpful. 

0 0
replied on May 14, 2015

This is a completely different Quick Fields session with one classification, but it is doing the same thing.

Here is the Document Class.

 

Here is what I get on the document.

 

0 0
View 3 previous replies
replied on May 18, 2015

Thanks for posting those. Let me make sure I understand what's happening:

For the example in this screenshot, Check Number and Customer ID fields are both incorrect? But the correct text is captured for those fields (using Zone OCR) when this document is used as a sample?

And for Customer Name, you didn't specify anything for that field in the configuration, but the "AE" shows up there somehow?

0 0
replied on May 18, 2015

That is correct. Customer ID should be "AE" and Customer Name should be blank.

0 0
replied on May 18, 2015

I see. Please open a support case with your reseller for this issue, and attach the session with the case. 

0 0
replied on May 19, 2015

A Support case was opened through your reseller. We will resume the troubleshooting process there, and once a resolution has been reached, we will update the thread here.

 

Regards

1 0
replied on May 28, 2015

Upon further investigation into the issue, it was determined that the user's session had the "Retrieve Fields" checked. This can be found under the blue nut/scan icon near the top left corner labeled "Configure the current scan source". Go into "Document Content" > "For each document:", and you should see "Retrieve Fields" underneath it as shown here,

 

 With it checked, this results in any value that is already present in the document to remain there after the Quick Fields processing phase. And as such, the user discovered that the scanner in which they were importing the files, was putting in field values when it should not have. Once that was addressed, the issue no longer persisted. 

 

So there are two ways to go about this,

1) If you need certain field values to be retained, then you can leave the "Retrieve Fields" checkbox, checked. However prior to the Quick Fields session, it should be confirmed that all the desired field values are correct, and that there is no additional user and/or workflow that is adding unwanted field values into the document prior to the Quick Fields session.

2) If no field values need to be retained, then that checkbox "Retrieve Fields" should be unchecked. This will safely ensure that no presents field values are transferred over into the Quick Field session.

 

Hope this helps.

1 0
replied on May 28, 2015

This was the case with my checks session however I get the same issue with my invoice session. I updated my VAR with this information. For the invoice session the 2 fields are blank when Quick Fields picks up the document. I can't uncheck the retrieve fields values box on the invoice session as there is information that the scanners input into a couple of fields.

0 0
replied on May 13, 2015

Yes, I do know they have been identified correctly, because when classed the vendor name is populated into a field. I will post a screen shot of the next one when it occurs.. 

1 0
replied on May 13, 2015

Hmm. So you've got a document class where the Account number field is set directly to "NoAccountNumber", instead of to a token? But documents that are identified as that class are showing something else in the Account number field?

Have you double checked that documents you're looking at have been identified correctly? Can you give an example of the text that ends up in the field in this situation?

1 0
replied on May 8, 2015

They are located in the page processing area. Also what is weird is I will have on some classifications where I have programmed the Account number to say "NoAccountNumber" as a fixed field, and still it will replace the information with other non-logic information

0 0
replied on May 7, 2015

Regarding the 'gibberish text' issue: Can you please elaborate on where in the processing your Zone OCR and Pattern Matching processes occur? For example, do you have a Zone OCR process and Pattern Matching both in First Page Identification? Are one or both of them of them in Pre-Classification Processing? Or Page Processing?

 

Regarding the splitting issue, does the hotkey for splitting work (CTRL + D)?

1 0
replied on May 4, 2015

Here is a screen shot. As you can see I selected the second page. There isn't a particular document that this happens with, it is random. I have tried sending back to Laserfiche and rescanning, but the same thing happens again. The only way I can resolve is to split it in Laserfiche and then reprocess through Quick Fields. 

0 0
replied on May 1, 2015

1. Can you post a screenshot of the session configuration? (make sure not to show any sensitive information)

2. Does this happen with any particular type of documents? Have you selected a page that isn't the first page before you try to click the 'split' option?

1 0
replied on May 1, 2015 • Show version history

1. I just checked all of my classifications and all of them have page 1 selected for both the Zone OCR and Pattern match. A lot of these that this happens with are one page invoices. 

2. The ability to not split the document is after I have assigned it to a document class. Sometimes it has been classed and I just need to split the document. 

0 0
replied on May 1, 2015

1. Is this happening only for multi-page documents? If the Zone OCR runs on all pages (as occurs in first page identification), your correct Zone OCR values are likely getting overwritten by text in the same location on subsequent pages. 

 

2. To be able to split the document when it is Unidentified, right click on it in the Document Revision pane, select 'Assign Document', and choose the correct Document Class. Once it is assigned to a class, you should be able to split the document. Then, if necessary, you can use 'Assign Document' again if the new document should be a different Document Class. 

1 0
You are not allowed to follow up in this post.

Sign in to reply to this post.

Details
Related Posts
Loading ...