Help with scatteredInterpolant: masking and meshgrid alternatives

Hello, thanks for reading this,

I am asking about ways to view a 3D point cloud as surfaces. This is a follow up to an earlier question: what I have is a 4 column text file denoting a point cloud with one column denoting data that I use for color, and three column entries for x y and z coordinates.

Here is the code I used, to get an idea:

https://gist.github.com/anonymous/8b3eab89bc708155900c

What I do right now (after looking online) is use scattered interpolation and apply the interpolation to a meshgrid, but there are problems with this approach. The problem is I am transitioning an irregular point cloud to a uniform mesh grid which I then surf to visualize. So, that being said, I can view a slice of the volume I have nicely, but the interpolation doesn't end at the boundaries of the volume: it goes to each end of the meshgrid I assign it to.

I was wondering: how can I create a similar meshgrid, but assign the surrounding region such that it isn't included when interpolating color? I think the problem arises when I try to create a meshgrid out of the interpolated data in line 23 of my code, and I think I can edit my code here to judge where to place NaN or actual data values, but I was wondering how I could do that properly.

I will attach a picture denoting what I mean as well.

Thanks for your advice.

Answer 1

[David Young]

A quick thought - would it help to switch off extrapolation? Use

F = scatteredInterpolant(X,Y,Z,C,'nearest','none');

[Brian]

This worked perfectly, thanks!

[David Young]

Oh good! You could accept the answer - flags up that it might be generally useful.