Alternate way to handle poor screen scraping data using Abby OCR engine

News Knowledge Base
1

Unable to get proper output when the user tries to read the data from an image using Abby OCR engine.

We can use “load Image” activity which will give us Image variable as output and we can use that as input for the “Abby OCR” Activity. The output will be proper.

 

The following scanning settings are best if you plan to perform OCR on the image:

  • Resolution: 300 dpi for font sizes of 10 points or larger and 400-600 dpi for font sizes of 9 points or smaller.
  • Scanning mode: Grayscale.


It is recommend scanning documents at 300 dpi for proper results.

  • For best recognition results, vertical and horizontal resolutions of the document must be the same.

Setting the resolution too high (over 600 dpi) slows down OCR. Increasing the resolution beyond this point does not yield substantially improved OCR results. 
Setting an extremely low resolution (less than 150 dpi) adversely affects OCR quality.

You may need to adjust the resolution of your images if:

  • The resolution of your image is less than 250 dpi or higher than 600 dpi.
  • If the image has a non-standard resolution. 
    Some faxes, for example, have a resolution of 204 by 96 dpi.

(These are the Recommendations provided by Abby Team)

1 Like