Surface plot from 3d scatter of a filled object

I have a data set that is a bunch of 3d coordinates. I have to reconstruct a 3d surface from it. I would use mesh or contour, but my points are not IN the surface, but rather some are in, while some are above. all I know is I have no points below that surface.

I tried the following code snippet and similar ones, and while it does get close, it has issues where the points are dense (I have a lot) because it tries to make a surface that hits every point.

! list XYZ are the scatter points
Xq = linspace(min(listX),max(listX),1000); ! I tried reducing the resolution here, and it helped a little,
Yq = linspace(min(listY),max(listY),1000); ! but it did not resolve the issue. 
[X,Y] = meshgrid(Xq,Yq);

F1 = scatteredInterpolant(listX(:),listY(:),listZ(:),"natural","none");
Z = F1(X,Y);

! I don't really know how to approach the problem, from a math standpoint. any hint or jumping off point would be much appreciated.

[Monica Roberts]

Sounds like you need the curve fitting toolbox, try "cftool" (or "curveFitter" in R2022a). If you don't have the curve fitting toolbox, I'd suggest a trial.

[Eric Ledieu]

thank you for the suggestion! I am trying the different fits. I'll put up the answer if I find it.

Answer 1

[Moksh]

Hi Eric,

I understand that you want to generate a 3d surface from a set of 3d coordinates with some of them being outside and inside the surface.

You can try using the “fit” function in the “Curve Fitting Toolbox” in MATLAB which creates a surface fit to the input data. You can specify the type of fit in the function along with the data provided according to the requirements.

Here is an example code for using this function:

% Generate random 3D coordinates
numPoints = 1000;
x = rand(numPoints, 1) * 10;
y = rand(numPoints, 1) * 10;
z = rand(numPoints, 1) * 10;

% Fit a surface to the random data
sf = fit([x, y], z, 'poly23');

% Plot the surface
plot(sf);

For more information about the “fit” function and which type of fit to use please refer to the following documentation:

• "fit" - https://www.mathworks.com/help/curvefit/fit.html
• "types" - https://www.mathworks.com/help/curvefit/list-of-library-models-for-curve-and-surface-fitting.html

Hope this information helps resolve the query

Best Regards,

Moksh Aggarwal