Check box recognition methods

Datacap employs optical mark recognition (OMR) to determine whether a check box option is selected.

There are two basic OMR techniques.

• OCR/A check box recognition method: This method is easy to set up and works well with non-dropout check boxes (where the check box outline remains on the page image). The method does not work as well with drop-out check box (where the outline drops out during scanning). The OCR/A recognition engine determines whether the specified region represents a selected check box (1) or a non-selected check box (0).

  Selected	Selected	Not selected

• Pixel threshold evaluation method: This method is more difficult to set up but is more reliable for drop-out check boxes. The method can also be used to read filled-in bubbles (O) on a response form. It calculates the percentage of black pixels within a specified zone and compares the result to a predetermined threshold value. For example, if the threshold is 20%, any OMR zone with more than 20% black pixels is considered selected (1). Any zone with 20% or less is considered not selected (0).

  > 20% black	> 20% black	<= 20% black

The field setup requirements are the same for both setup check box recognition techniques.