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

Question

Question

Zone OCR in Quickfields

asked on February 27, 2014

 I have a couple of reports that contain a list of SSN's and names. Report can have contain a variable list of SSN/Names from 1 - 20 per page. I can setup Zone OCR within the report when scanned with the intent of full text search on these fields. Has anyone attempted to Zone OCR a variable results scanned report and if yes what is the success rate picking up all the variables for full text search? Secondly, is there anyway to verify that all variables were OCR'd and if not is there a way to add them.

0 0

Answer

APPROVED ANSWER
replied on February 27, 2014

Zone OCR data can be used to populate field values or set the document name and path. If you're looking for full text search specifically, you want to OCR the entire page.

 

If your names are listed in a table (in other words if there's interference from other text), you can make a Zone OCR big enough to handle 20 of them and return a multi-value token that you could write to a field. You could use a token function on that multi-value token to count the values and write it to a different field. But that would assume you know ahead of time how many you expect. And if any are missed, then the user would have to type them into the field.

0 0

Replies

replied on February 27, 2014

Can you clarify a bit better.

 

It sounds like you want to do a few different things, if you can break them out clearly I will try my best to point you in the right direction. 

0 0
replied on February 27, 2014

Kenneth, we are a government agency and administer benefits for several hundred thousand teachers. We interact with several other agencies that administer benefits such as insurance and payroll. We get a report monthly from another agency that provides EFT changes such as accounts and routing numbers. We have a separate process to update that information. What we need to do is track all correspondence/activity that occurs on each members account. In the past they would just photocopy x number of copies of the report and put a copy in each members record. Now that we have switched to imaging, they still want a copy stored for each member, but that would be extremely redundant. We could create shortcuts with links for each member and link to the primary. That would require mass data entry. The thought was if we OCR'd the report that staff could use Search to find all documents related to the SSN or name on the report. The report is rarely accessed. I am mainly looking to see if others have done something similar and hoe effective it was. Thanks!

0 0
replied on February 27, 2014

I'm not positive I understand this correctly, but as I get it there can be any number of SSN's on a page and you want to grab them all?

Why not instead of a Zone OCR just OCR the whole thing, then use a pattern matching process to pull out all of them.

 

0 0
replied on February 27, 2014

Yep. I've found that Zone OCR is only useful when the structure of the document is 100% predictable. If you have variable positions or numbers of fields then full page OCR is the better option.

0 0
replied on February 28, 2014

Devin, we have had similar problems with Zone OCR.  Can you or other users tell me based on your experience if I use full page OCR how effective OCR is in capturing all document text for full text search?

 

Thanks

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

Sign in to reply to this post.