undefined | Better HN

0 pointsuSoldering2y ago0 comments

Making a machine to automagically remove a tiny bit of material and image the result over and over would be easy for me. The image processing to take the stack of 3D sequential images and automagically process them into a netlist is well beyond my programming capabilities. If anyone thinks they could do this, contact me.

0 comments

3 comments · 1 top-level

Keyframe2y ago· 2 in thread

Not gonna pretend I have the solution, but it sounds like most of the groundwork for that has been laid out in medical imagery already. CT scans, combined into volumes, identifying structures..

alright25652y ago

That's what I was thinking, but now I'm pretty sure it doesn't even need crazy algorithms like that.

1. align the image stack. not trivial, but a common task.

2. take several cross-sections, in both dimensions, and have a human draw a line along a specific layer line

3. linearly interpolate these lines into a surface.

4. for each pixel in each output layer, set the value to layers[l + offset][x][y], where the offset was calculated in step 3.

willis9362y ago

Steps 3 and 4 seem like a largely solved problem using blob detection in computer vision libraries. The last step is in plated via detection. If no microvias are present this seems pretty easy: just look for circles. If microvias are present then you need to check for plating at each layer for each hole below a certain size. That seems difficult if the sanding process does not result in very clean features.

j / k navigate · click thread line to collapse