I have sonar image which is basically a point cloud (x, y, z) of an irregular
underground cone-shaped structure. I am trying to create either (1) a closed 3D
mesh with "equally spaced" mesh grid, or (2) a smooth closed surface, for
instance by calculating connected localized regression surfaces. My Z
(vertical) component is equally spaced, but X and Y consist of 128 points
around the perimeter, regardless of diameter, which makes the grid size small
where the diameter is small, and large where it is large. I'm looking for
relatively equal grids, so I could use regression or interpolation to add
points where needed, or conversely remove them, hence the idea of a localized
surface regression. I've tried plot3D package, RGL, and a couple other packages
including alphashape3d. I have not tried plotly or rms. So much time invested I
thought I would ask for advice, if anyone knows a package or technique or can
clue which direction to go. Thanks,
Confidentiality Notice: This E-mail and any attachments ...{{dropped:12}}
______________________________________________
[email protected] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
