Skip to main content
Kofax

OCR zone fails to read text and populate Index Field

3019266

Question / Problem: 

Why does a drawn OCR zone saved in a Recognition Template fail to read text which appears to be in the same location on the page and use that text to populate the associated Index Field?

Answer / Solution: 

The Kofax Express Help Setting Up a Recognition Zone Template section states:

"if you expect to reuse recognition zones indefinitely, use the Template Mode technique, which is useful for indexing information that appears consistently in the same location from one document to another."

If the OCR zone is drawn too tightly around the text, slight differences in scanned image size or the position of the printed text on the physical page can cause some or all the text to fall outside the bounds of the zone.

Example:

Image 000000001.tif has the size of 1720x2220 pixels compared with

image 000000002.tif has the size of 1713x2212 pixels.

They have 7 pixels different in height, and 2 pixels different in width.

 

If any portion of the text falls outside of the OCR zone, that portion of the text will not be read.


When indexing in Kofax Express, the user can manually draw an OCR zone around any text that falls outside the zone defined in the template.

1. Select the index field to populate with the OCR results

2. Press Ctrl + Shift + D to enable Draw OCR Zone

3. Draw an OCR zone around the text to be recognized.

4. The OCR results will populate the index field.

 

Applies to:  

Product Version
Express 3.2

 

 

  • Was this article helpful?