Local Plane Fitting/Alternative Normals Metrics

[mb1181]

I am wondering if CloudCompare has the capability to generate other scalar field values from normal calculations besides the available Dip/Dip Direction scalar field results.

If I understand how normals are being computed correctly, I am interested in getting the absolute height of the normal vector's origin (i.e. the absolute height of the center point of the plane being fit to generate the normal vector). The way I understand dip and dip direction, these values only reference the alignment of the normal vector and not the position.

My end goal is to use this as a method for classifying local/micro deviations/bumps in scan data from the larger macro surface. Currently using a difference between "dip" values between different fitting radii has not seemed to produce the results I was looking for, I believe largely because the relative alignments of the micro and macro surface could be similar while being offset from one another positionally.

Maybe this is not related to the way normals are calculated, and if so I would be interested in finding another approach to get a similar result.

[daniel]

It's quite similar to the way 'roughness' is computed: see https://www.cloudcompare.org/doc/wiki/i ... /Roughness

<< Roughness estimation is very... simple: for each point, the 'roughness' value is equal to the distance between this point and the best fitting plane computed on its nearest neighbors. >>

Wouldn't it be sufficient?

[mb1181]

Yes. The roughness computation is very helpful, accomplishing what I am looking for. I was not familiar with it, thank you for pointing it out.

The only question I have is if within the roughness calculation if it is possible to select different plane calculation methods (similar to computing normals with plane, quadric, and triangulation surface modeling methods).

Thank you

[daniel]

Ah no, especially since with the other methods it's not planes... The roughness only uses a least squares fitting plane.