Skip to main content

If image exists

Sophie avatar
Written by Sophie
Updated this week

Definition and Usage

Check if a specified image is found on screen. This command allows you to verify whether a particular image exists or does not exist on the entire screen, a specific window, or the currently active window.


Parameter Values

Input parameters

Parameter Name

Description

Possible Values

Required

Options / Notes

Search scope

Specify the scope to search for the target image

Entire screen, Specific window, Currently active window

Yes

If "Specific window" is selected, you must provide a window object

Window

Select a window object that was obtained earlier

-

Yes, if "Specific window" is selected

-

Select image

The image to locate on the screen or within the selected window

-

Yes

Use the Capture button to select an image

If image

Check for existence or absence of the selected image

Exists, Does not exist

Yes

-

Advanced settings

Parameter Name

Description

Possible Values

Required

Options / Notes

Match mode

Color matching uses full color comparison; grayscale matching compares grayscale images and is faster

Color matching, Grayscale matching

No

Default is Color matching

Similarity (%)

Sets the similarity threshold (%) for image matching

0-100

No

Recommended: 80%

Variables produced

This action doesn't produce any variables.


Using Variables in Conditions

When you see {x} in fields like Similarity (%), you can insert variables created earlier in your workflow. This is useful when you want to dynamically set the similarity threshold based on different scenarios or previous operations.


Notes

  • The image matching precision depends on the similarity threshold - higher values require more precise matches.

  • Screen resolution, color settings, and scaling may affect image recognition results.

  • For better reliability, capture images with distinctive features rather than plain or common elements.

  • Image detection is sensitive to visual changes; consider using a lower similarity threshold if the target image might vary slightly.

Did this answer your question?