search cancel

Form Recognition is not giving reliable results

book

Article ID: 175881

calendar_today

Updated On:

Products

Data Loss Prevention Enforce

Issue/Introduction

You have created a PDF Form Recognition archive document as a reference, but when testing DLP recognition capabilities you are getting inconsistent results

Environment

14.x, 15.x

Cause

There are multiple possible reasons that a PDF reference-document may be unreliably detected in DLP:

  1. It is not saved in a high enough resolution
  2. Its content is not evenly distributed across the page
  3. The document background is too dark
  4. Electronically fillable fields are not in the AcroForm format

 

Resolution

  1. Use a third party site such as Pixel calculator to work out the DPI of your PDF document by entering the resolution from the document's properties window into the pixel values on the calculator site. If necessary recreate your PDF file with a higher resolution and re-archive it.
  2. Always view the whole page before creating the archive to ensure content is evenly distributed
  3. Avoid any shading for backgrounds in Form Recognition documents. Keep it black on white for best results
  4. Ensure your PDF creation tool is set to use AcroForm format if you have any electronically modifiable fields. 

 

For detailed information see the Administrator's Guide section 'Preparing a Form Recognition Gallery Archive'