algorithmic modeling for Rhino
while I'm analyzing curvature, I would like to find the inflection points of a curve. i could filter them out evaluating the largest curvature radii..
but is there a way to get them more exact way. i remember something from school like if the 2nd derivation changes from positive to negative there is an inflection points. but why are the derivations in gh vectors?
would be very thankful if somebody could enlight me.
wouldn't it be nice to have a component for that in future versions?
Here's what seems at first glance like a pretty good approximator. Testing might prove otherwise. I culled out the end segments as the information near the end points seemed questionable (or perhaps I wasn't understanding it correctly). Adjust the minima/maxima slider to get the least number of points that 'look correct' for the shape of the curve. Adjust the resolution slider for the inflection points in a similar fashion. If you get too fine or too coarse with it it won't give nice output.
Thank you Chris, it works very well.
thank you chris,
just checked it, seems to work pretty well.
but how is it with an non planar curves? they can't really have inflection points right?
found this discussion and experimented but still i dont really understand the signification and connection of the torsion and the derivations of the crv..
Here's another approach for finding inflection points. I just made something similar and then happened upon this post.
The intention for mine was to find all the local curvature minimums and maximums of a curve. Inflection points happen to be the point of least curvature. Results are as exact as the resolution you specify. Planar orientation doesn't matter. Works just as well on non-planar curves.