I want to recognize a couple of fields in a document but the documents (although the same form) Might be scanned in different resolutions or maybe skewed.. Does the cloud recognition engine make corrections?

asked 29 Jan '14, 04:51

Francisco%20Tirado's gravatar image

Francisco Ti...
111


No, unfortunately it does not.


If the fields you need contain the common (not handprinted) text only, you can recognize full page, export all the text with its coordinates to xml and find the text you need.


If the fields you need contain handprinted text (or checkmarks - in these cases it is necessary to specify exact fields coordinates before recognition):

You can define field coordinates using coordinates of another object (it could be symbols like "@", the best results are achieved when it is a barcode or a couple of words; to let short text fragments be recognized, use the textExtraction profile).

So, the scenario in this case is:

  1. Recognize full page using the processImage method, export the result to xml.

  2. Get the coordinates of the object you know from this xml.

  3. Using this coordinates calculate the coordinates of the field you need.

  4. Recognize the necessary field using field-level recognition.

It will be billed as one page (you can recognize fields on an image that was already recognized for free).

link
This answer is marked "community wiki".

answered 29 Jan '14, 14:13

Anastasia%20Galimova's gravatar image

Anastasia Ga... ♦♦
790112

wikified 29 Jan '14, 14:13

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Tags:

×3

Asked: 29 Jan '14, 04:51

Seen: 488 times

Last updated: 29 Jan '14, 14:13

© 2016 ABBYY. All rights Reserved. www.ABBYY.com | Privacy Policy | Legal