How can I perform edge gluing with mech vision on version 1.7.2?
I want the trajectory to follow a certain direction for a unique shape.
For example:
How can I perform edge gluing with mech vision on version 1.7.2?
I want the trajectory to follow a certain direction for a unique shape.
For example:
Mech-Vision has this Step for the function you need. But please contact the Mech-Mind technical support for your company to test and deploy it.
Hi, as 2nd image showned above, Mech-Vision also has this step [Generate Path Given Depth Map]. It can generate path like the screenshot that I attached. And you can see the test result from the 1st image.
@Eric What version of Mech Vision are you using? Because Iâm using version 1.7.2 and cannot find the âGenerate Path Given Depth Mapâ step.
Thanks @Eric I can use it now but, I want to ask if you could show the result obtained for âCalc Mask for Highest Layerâ and âPoint Cloud Pre-Processingâ? Currently I can only get 2 poses on a flat-triangle shape, and they are not in the corners.
Click right button, and show more parameters.
Try to minimize the Erosion Kernel Size and Sample Distance.
Double click the line and the debug output will be shown in the top right corner.
Hello Mr. Eric
I am working on a Robot project that injects glue into the workpiece similar to your old project. I have thanks to the Mech-Mind NaNo Camera support, Vision and Viz 2.0.0 software. Iâm having trouble creating a project on Vision software. I tried to find blocks on version 2.0.0 to match your version 1.7 but without success. Could you please give me your vision file? Or can you help me get the blocks on version 2.0.0?
We look forward to your agreement, thank you
friendly
Mr Namtranhieu
Hi shawn.Zhou
I followed the step you instructed, I successfully achieved the âcreate path for depth mapâ block. As for the âpoint cloud preprocessingâ block and the âcalc mask for the highest layerâ block, I canât find it in the âstep libraryâ. How can I get those two blocks out on Vision 2.0.0?
I think the âpoint cloud preprocessingâ block is the â3D target object recognitionâ block.
Hope you reply
Mr Namtranhieu
Those procedures (a combination of multiple steps) have been removed in Vision 2.0.0
You can check the list here: Mech-Vision 2.0 Release Notes
In the scenario you described, the deleted procedures are not indispensable. You can achieve the same effect using other stepsďźsuch like Point FilterăGet Highest Layer CloudsăPoint Cloud Clustering
The AI Assistant is now available for staff.
Come and try it out!
I am here to assist with your technical questions about Mech-Mind.