1. Zebra Aurora Focus User Guide
  2. Using Machine Vision Tools
  3. Identification Tools
  4. Using Accelerated Tools
  5. Deep Learning OCR
  • Zebra Aurora Focus User Guide
Deep Learning OCR

Deep Learning OCR

The Deep Learning Optical Character Recognition (OCR) tool reads text from images using Deep Learning.
This tool locates and recognizes characters without additional training:
  • Horizontally-oriented
  • Height between 85% and 115% of Char Height (in pixels)
  • Contain Latin letters (upper case or lower case), digits, or one of: !#$%&()*+,-./:;<=>?@[]^_`{|}~"'\€£¥
Use the Deep Learning OCR region of interest to limit the analyzed area, which can lead to improved performance. Moreover, it may be used to adjust text that is not displayed horizontally.
Deep Learning OCR Settings
Setting
Description
Minimum Confidence
The Minimum Confidence parameter may be used to change a character's minimum score. By default, this threshold is set to 80%.
Adjusting this value downward can help include decodes of slightly lesser confidence to get the desired output. For example, the above example is too strict and does not decode various characters. Lowering this value from 99 to 98 yields a favorable result.
Average Character Height
The Average Character Height should be set to the average height of characters (specifically, capital letters) in the analyzed area. For example, if an image contains two kinds of characters: one is 24 pixels high and the second is 40 pixels high, the Average Character Height should be set to 32, irrespective of the number of characters of each kind.
Use the slider to select a value that is close to the average pixel height of text that is trying to be decoded. Performance improves when this value is closer to the average character height (in pixels).
  • Minimum Value: 8
  • Default Value: 25
  • Maximum Value: 200
Character Spacing
Distance between characters is denoted as a fraction of CharHeight.
Text Color
Set a required polarity for a character to be returned.
  • Bright: Only characters with contrast greater than Contrast Threshold are returned.
  • Dark: Only characters with contrast lower than Contrast Threshold are returned.
  • Any: Only characters with contrast lower than Contrast Threshold or greater than Contrast Threshold is returned.
Default Value: Any
Contrast Threshold
Sets a threshold for a contrast of found characters.
Default value: 0
Character Width Scale Percentage
Scales image width by the given factor (%).
  • Minimum Value: 10%
  • Default Value: 100%
  • Maximum Value: 1000%
Character Range
Enable Character Range to limit the set of recognized characters.
This string must be formatted according to the following rules:
  • Allowed characters must be separated with commas.
  • For ease of use, a continuous range of letters or digits may be written as starting_character-ending_character, for example: A-Z or 1-6.
  • Comma and backslash have to be prepended with a backslash.
For example, Character Range equal to A-F,g-o,0-9,X,Y,Z,-,\\,\, results in recognizing only ABCDEFXYZghijklmno0123456789-\, characters.
Character Range
Provides the option to limit the set of desired characters when returning the result.
String Match
Defines a string that must be contained within the decoded OCR output for the tool to pass (for example, substring). By default, the tool passes if the match string is contained in the resulting OCR output. This allows additional characters to be added before and after the string.
Enable the RegEx checkbox and use the syntax
^<stringToExactMatch$
 to implement the exact match behavior.
RegEx can also be used for complex string match logic. Refer to the RegEx section for more details.
Timeout
Sets a time after which the tool fails.
Invert Pass/Fail Result
Flips the results of this tool.