Question

How can we bring all results from a Zone OCR across all pages in a document into a single multi-value field?

Quick Fields Capture

Updated June 22, 2015

asked on June 18, 2015

Working with processing batches of tax notices that have varying numbers of pages in each document. We want to capture all the roll numbers in that column across each page of the notice into a single list of roll numbers into a "Roll Number" multi-value field. With pattern matching i can easily grab all the values on a pages into that multi-value field but I want to append all the entries from each page into the same field. I'm looking at the token accumulator in post processing but the documentation appears like it will not gather all values in a multi-value field for each page.

How can I accomplish this?

0 0

Answer

SELECTED ANSWER

replied on June 19, 2015

That looks the same.

I see the behavior you're seeing if Zone OCR is set to page 1 instead of all pages.

Could you move both Zone OCR and Token Accumulator in Page Processing and try again?

1 0

Replies

replied on June 19, 2015

I have Zone OCR set to read all pages, single line False and create multi-value token True. Token Accumulator is set to all pages as well, collecting the Zone OCR token. I'm not getting any duplicates for single or multi-page documents. Could you confirm your Zone OCR configuration?

2 0

replied on June 19, 2015

We have a Token Accumulator activity in Quick Fields 9:

1 0

replied on June 19, 2015

So you are confirming that we should expect the token accumulator to collect the multi-value token from Zone OCR for each page into a single multi-value token? I ask because the documentation seems to say it won't with multi-value tokens and so far our testing seems to confirm it. However, if you believe it's supposed to work I will go back and work with it some more.

1 0

replied on June 19, 2015

Ok I had a look at this remotely with the customer. I confirmed: The token we're using to accumulate is the multi-value token from the Zone OCR. That Zone OCR is running on all pages of the document. Here's exactly the result we're getting:

WITHOUT using the token accumulator the Zone OCR token just keeps the last page. It looks like it overwrites the token for every page of the document so the result is:

Page 1 Roll Number column:

111

222

333

Page 2 Roll Number column:

444

555

Zone OCR token:

444

555

I use the Token Accumulator in post processing and set post processing to run before storing the document. I use the Zone OCR token. The result I get for the same sample data as above is:

Token Accumulator data:

444

555

444

555

If there's a single page document (using the same sample data as above) the result is:

Token Accumulator data:

111

222

333

111

222

333

1 0

replied on June 18, 2015

Example: Page 1 Roll Number column:

111

222

333

Page 2 Roll Number column:

444

I want the Roll number field to read:

111

222

333

444

Notices have 1 or more pages.

0 0

replied on June 19, 2015

0 0

SELECTED ANSWER

replied on June 19, 2015

That looks the same.

I see the behavior you're seeing if Zone OCR is set to page 1 instead of all pages.

Could you move both Zone OCR and Token Accumulator in Page Processing and try again?

1 0

replied on June 22, 2015

Moving the Accumulators into Page Processing worked. Thank you!

1 0

You are not allowed to follow up in this post.

Question

Question

How can we bring all results from a Zone OCR across all pages in a document into a single multi-value field?

Answer

Replies

Sign in to reply to this post.