this is my first post, but as a silent reader I learned a lot, so first of all: Thank you, you are awesome!

I now have a problem, which I don't know if it belongs here, but I'll give it a shot:

For scientific researches my group did a CT scan of a tube filled with hundreds of hexagonial prisms (a so called 'fixed bed'). The goal is to compare this 'real world' filling with filling simulations done with Blender (works great btw ;) ). Unfortunately, the .stl-file provided by the CT scan lacks of 'resolution' (see the attached image): You can see the different pellets, but when you look at the contact points/edges/faces, you see that the surfaces bake together, which is not what we want. We want a closed surface for every single pellet (with empyt spaces between). A CT scanner with higher resolution might solve this problem, but I am curious if there is another solution based on software and smart people.

So long story short: Is there an algorithm/ workflow in Blender which can detect shapes and isolate them from each other? Or do you know another software which is more suitable?

I am grateful for every single idea! Cheers,

LeoCT scan: Fixed Bed

  • This is more of a scicomp.stackexchange.com cs.stackexchange.com/help/on-topic datascience.stackexchange.com question. My thought is to use an iterative algorithm depending on the data source(Point-cloud, mesh, CT-Slices?) .Find best fitting hexagonal prism, (get orientation, position, scale? remove contained geometry find next best fitting prism). I would guess, that working of the slices themselves is easiest, as the intersection between the prism and the slice will probably give you a characteristic hexagonal shape from which you can guess orientation of the prism. – miceterminator Sep 14 at 12:56
  • Can you provide a wireframe or edit mode image. Interesting q, some iterative routine to reduce to edges of two known lengths, and 2 known face shapes. Can you add a sample file blend-exchange.giantcowfilms.com – batFINGER Sep 14 at 15:23

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.