plot 3D point data

I have matrix A=[x y z] with 2112 point data . Please see attachment file, and figure:

How can I plot the surface using that data as the illustration figure?

Answer 1

[Rik]

You should be able to use the surf function. The only problem you have is interpolating the data to a normal grid that surf understands (and fill the rest with NaN).

The code below assumes your text file is stored in a 2112 by 3 array named data

x=data(:,1);
y=data(:,2);
z=data(:,3);
SamplePerDim=500;

X=linspace(min(x),max(x),SamplePerDim);
Y=linspace(min(y),max(y),SamplePerDim);
[X,Y]=ndgrid(X,Y);
F=scatteredInterpolant(x,y,z,'linear','none');
Z=F(X,Y);

figure(1),clf(1)
surf(X,Y,Z,'EdgeColor','none')
view(-160,80)

[ha ha]

@Rik Wisselink. Can you show me the detail code?

P/s: I also tried with surf function. But I need the data were plotted in the original grid(not normal grid).

[Rik]

See edited answer

[ha ha]

I try your code. It's good. May i ask you a favor? What is the best number of "SamplePerDim" ? And, How to choose automatically the best number of "SamplePerDim" based on number of the input data? Since when I try with small value of "SamplePerDim=100". The result figure is not good

[Rik]

As always, you're welcome.

I don't really know how to pick the resolution automatically. That will depend on the smoothness of your data and how much data you have. Ideally you would 2000, but that makes the plot very slow. You could use a fancy formula to estimate the best value, but because that is tricky I chose to hard code the value.