OCR(Optical Character Recognition)

It reads the text information on the screen and provides it as a screen element. Currently, the characters that can be read are Korean, English, numbers, and special characters.

OCR is recognized word by word by default. If you need to use multiple word sentences as a single screen element, you can use multiple words as a single screen element by selecting multiple words through drag & drop.

OCR Result Inspector

Specific information about the words recognized by OCR can be found by clicking "Inspector" at the bottom of the screen analysis panel.

Click "Inspector" to display a list of recognized words in a table format.

  1. TEXT: The recognized word content is printed.

  2. BOX: On-screen location information (coordinates) of the word is output.

Clicking an item in the Inspector table activates the yellow selection in the screen analysis results.

Conversely, when you click a specific screen element in the screen analysis results, the items in the Inspector table become yellow.

OCR Screen Element Attribute

You can set various OCR properties by clicking on the screen element in the step using OCR.

See the link below for instructions on how to access the OCR screen element attributes.

Set Step Properties

  1. Text: You can set comparison conditions when testing with character information on screen elements.

    • = : Check that the text and value of the screen element are the same.

    • *= : Checks whether the value set to value is included in the text of the screen element.

    • ^= : Check that the value set to value matches the beginning of the text on the screen element.

    • $= : Check that the value set to value matches the end of the text on the screen element.

    • match: Use regular expressions. Check that the text of the screen element matches the regular expression.

    Reference: https://docs.python.org/3/library/re.html#re.match

    • search: Use regular expressions. Check that the text of the screen element contains a regular expression.

    Reference: https://docs.python.org/3/library/re.html#re.search

  2. Text Similarity(threshold):

    Indicates string similarity (available only when comparison condition is '=') You can set how identical the character information must be at the time of scenario creation and at the time of test execution to pass through. Due to the nature of OCR, recognition of the same string may result in some misrecognition depending on the resolution and other environments, in which case similarity can be adjusted. The higher the value, the higher the comparison accuracy. - Default value: 0.8 - Minimum value: 0 - Maximum value: 1

  3. Case Sensitive: For alphabetic characters, you can set case sensitive.

  4. Selector: Specifies which element to use when multiple elements match the screen element during the test. It can be set to an integer value other than zero. (Note: If you enter '-1', use the last element.)

PreviousOD(Object Detection)NextCrop Image

Last updated