Contribution to the factory io project 122 (Orient boxes according to their label using machine vision)

Shared Projects
1

Hello,

Let’s look at another variation of the previous contributions. As you can see, a few simple changes are all it takes to create another unique project online, using computer vision. This can’t be done with Factory IO as it is because it can’t read which side the label is on. In fact, it can’t be done with any existing simulator that I know of—I repeat, that I know of.

As you can see, FACTORY IO replaces the expensive prototypes currently used in class, which are limited to what the manufacturer wants to sell and not to what the student needs to learn. In other words, with a prototype, it’s just the manufacturer’s camera, period. Without a prototype, any camera will do. Therefore, the student doesn’t care where the image comes from; they’ll be prepared for anything that comes up. With a prototype, they won’t be.

I’ll share the scene later, in case it needs adjustments for some class… but if anyone wants it now, let me know and I’ll share it.

Regards

2

Hi:

Now it’s ready to share the scene. As you can see, it needs some modifications, so I’m not sharing it until I’m sure everything is working correctly.

See the general machine settings tests. All that’s left is to readjust the values ​​and we’ll finish the project. Keep in mind that you might not be able to do it as quickly because you’re missing some of the things I have, but don’t get discouraged if you don’t get it right the first time. It didn’t work for me either…

Regards.

3

Hello:

If everything went well after adjusting the tolerance at the input of the boxes, as you know, these systems work between two values, in this case the X coordinate. The result is this video.

I hope to see it soon online, created by someone who’s willing to try it.

Regards