Image Recognition RPA

Certified Senior Developer

It is regarding image recognition I store image of 8,+,5,= so as per pdf it should be excused in calculator and answer would be 13 but it is not executing and that why I am not able to complete this lesson

Facing the issue after completing the all steps provided by Appian Document

Anyone Guide me how can we resolve this....

  Discussion posts and replies are publicly visible

  • You might need to add a new image. Try selecting the eye icon in the Support files to verify the image is an '8'. If it is not, then replace this image with another. Also, if the image looks like an '8', you might try adjusting the Pixel tolerance or Color tolerance - for example to 'Very low' and test again. This is an example of the image that worked for me using the win32calc.exe calculator. 

  • 0
    Certified Senior Developer
    in reply to Randall Richeson

    Disappointed

    Facing the same issue

    I configured values below image

    and added images as like you said

    But not working

    guide me @ how can i resolve this issue...

  • Try increasing the Pixel and Color tolerance to higher values like Low or Medium. I have found that sometimes this works. If it does not, try creating a new robotic task and use the agent 'Create Support Image' option to create new images for it. Make sure you are capturing the images on the same host machine the robot will run on. I reproduced this same error with an older robotic task, but after creating a new robotic task with new images it completed successfully with an exact match for Pixel and Color tolerance. 

  • Hi  

    Setting Up Variables:

    1. Create a variable called "calculatorResult."

    Opening the Calculator Application:

    1. In the "Open Application" activity:
      • Set the executable name to "calc.exe."
      • Use "Find by Window Title" and set the expected window title to "Calculator."

    Creating Support Images:

    1. Manually open the calculator in the Appian RPA agent.
    2. From the Appian RPA agent menu in the taskbar, select "Create Support Image."

    Support Image Marking:

    1. Mark the new image for the number 8, save it, rename it, and click "Save Image."
    2. Repeat for the "+" icon, saving images accordingly.
    3. Similarly, save images for known values like 5 and "="

    Configuring Windows:

    1. For the "Eight" configuration window:

      • Select "Full Screen" in the search section.
      • Set the max wait time to 2 seconds.
      • Choose "Exact Match" for pixel and color tolerance.
      • Enable mouse actions.
      • For the location image, select "Throw Exception if Not Found."
      • After completion, set "Wait Before Executing Next Actions" to 4 seconds.
    2. For the "Plus" configuration window:

      • Set "Don't Wait" for wait time.
      • Choose "Exact Match" for pixel and color tolerance.
      • Enable mouse actions.
      • For the location image, select "Throw Exception if Not Found."
      • Repeat the same for the "Five" value.

    Keyboard Shortcuts:

    1. Use the "Press Keyboard Shortcuts" activity.
      • Change the display name to "Copy the Result."
      • In modifier keys, select "Ctrl."
      • In additional keys, select "Characters" and provide "c."
      • In interaction, select "Press and Release Keys."

    Getting Clipboard Value:

    1. Use the "Get Clipboard Value" activity.
      • In the configuration output section:
        • Select "Is Stored As" in the dropdown.
        • In the target, provide the variable created, i.e., "calculatorResult."
      • After completion, select the "Take a screenshot" option.

    Follow these steps carefully, and you should get the correct result. Refer to the accompanying screenshots for a better understanding.