CODEWITHSUNDEEP
machine-learningmeshfeature-extractionpoint-clouds
pacc
pacc

Reputation: 1

3D machining part Feature Recognition (Point Cloud, Mesh)

I have a machining part (.STL) and want to recognize (and extract) it's machining features. Some of the features are simple but some are more complex that's why I think a machine learning approach would be fitting because I can't describe the feature mathematically.

There is FeatureNet which basically does this job expect that it can't recognize multiple features and the code doesn't work as expected.

I also know of AAGNet which does what I want but it uses .STEP as input but I have a mesh (or point cloud if I convert it).

Since there are a lot more point clouds repositories I thought that I could maybe use them to solve my problem. Is something like FPFH the right direction or am I on the wrong path?

If I would use a machine learning approach I can easily create a labeled dataset.

Upvotes: -1

Views: 71

Answers (1)

el_grezeq
el_grezeq

Reputation: 133

Machining feature recognition is a still an active and quite narrow research field so I'd be surprised if you get any out-of-the-box solution here. The best you can do IMO is to keep trying out solutions from the literature and see how they work for you.

You could indeed get a pointcloud by sampling your original mesh, compute local descriptors and use them to train a classifier. But there is no guarantee that the local geometry around a point is enough to recognize a machining feature.

The solution described in the paper Freeform Machining Features: New Concepts and Classification uses a mesh as input geometry and combines differential geometry and graph theory to classify machining features. Maybe worth a try?

Upvotes: 0

Related Questions