![]() |
| |||||||
Testing Anywhere Post messages and questions related to Testing Anywhere here. |
![]() |
| | LinkBack | Thread Tools | Display Modes |
| |||
|
Hi, just creating a test with an image recognition checkpoint, that fails even at 1% match. This doesn't make sense. The image is compared to itself, not another image, so should pass, 100% of the time, under 100% match. Any ideas what is happening? Screen shot attached. |
| |||
|
Hello KMuller, This is not the normal behavior. We request you to please provide further more details regarding the Test. Also let us know, are you trying to capture the image from any website or any application? If you can forward the Test file (.tamn) we would like to have a look at the image captured and suggest you right solutions. You can forward the test at support@automationanywhere.com Thanks and Regards, |
| |||
|
I also removed my image recognition! what can I do to make it back. I don't know how to back it again. pleas help me on how make it again. thank you! from: |
| |||
|
Hello Kenneth21, To make the Image Recognition command back you will have to add the Image Recognition command from the command list to the task in which you want to add the Image Recognition. After opening the Image Recognition command, select the window title of the window on which you want Image Recognition to work. Image 1 is the main image which you will have to capture and image 2 is the image which you will be searching in image 1. For example: If we use Image Recognition on Calculator, then select window title as Calculator by keeping the calculator open. Now, Image 1 is the whole calculator window and Image 2 is the button 1 in the calculator window. Image Recognition will now click on the button 1 in the calculator so we can say that Image Recognition will search the button 1 in the calculator and will perform a click on it. Thanks & Regards |
| |||
|
TA's suggestion was to increase the size of the captured image. It seems to have worked (for now), but I have two problems with this: 1) This is completely undocumented. There is no 'minimum image recognition size' in any sales or help documentation. 2) The specific image I am using in this case is already 95% black with two small images in the middle. If I increase the image size, there will be more black, and more chance the checkpoint will pass, when it should fail. |
| |||
|
Hi Karl, When you use the 'Image Recognition' command, the window that pops-up to select the images to be captured has an option called - 'How to use?' on the top right hand corner. If you click on the link it will give you documented instructions on how to use the command. Let us know if that is something you were looking for. |
![]() |
| Thread Tools | |
| Display Modes | |
| |