Transformation of non-euclidean geometry to euclidean projection

Hello there! i have this little problem to solve: i have a 2d-projection of two perpendicular lines on a sphere surface, i have no information on the generating sphere however. those lines are arches of a circonferences, but i need some kind of geometric transformation to revert them to the original straight perpendicular lines (of wich i have measures and true form).

There is some scripts to do this magic, or can you suggest some way to do it?

thanks!

P.s. i’m adding an image to be more clear:


In the image there are several spheric projections (in green) the one on the center is in “true form” due to being centered with the axis of the sphere. the red perpendicular lines are the original lines.

I have to find a way to transform the green lines (like the one in the top left coner) back to the straight perpendicular red lines. with this transformation i hope that would be possible to transform back to “true form” also any shape that would be contained into the the green lines.

you should post a file.

I can’t really make sense of your description.

how can the length of the curve segments be identical, if one is on a plane (red original) and the other a projection on a sphere (if projected in the plane of the original curve)?

you can recreate the sphere by using a 4 point sphere on any of the projected green curves…

but I don’t understand what you are trying to reconstruct from there.

I believe the green curves are projections of the spherical surface curves on to a flat plane

which means the 4 points sphere isn’t suitable.

So the desire is to recreate the spherical surface curves (without knowing the sphere radius!) and thence to recreate the original planar curves.

Hi @The_G,

Questions:

  1. Did the original planar red lines come from a plane that is parallel to the plane on which the green lines lie?

  2. Do you know the distance from the centre of the sphere to the red lines plane?

  3. Are your example green lines a contiguous set, spaced as shown, projected there from the same sphere? Or six independent examples which you have laid out side by side on the page?

Regards
Jeremy

Hello- if the green curves are non-planar, lying on a sphere, you should be able to approximate the sphere by sampling points on the curves (Divide command) and then Sphere > FitPoints.

-Pascal