smooth a surface plot

I have an array of data which I'm plotting using the surface function. I'd like to smooth the data but interp2 doesn't work since I don't have a mesh (I'm plotting data, not a function). Is there a way I can smooth this?

My code isn't much to look at:

surface = surf(P1,P2,L);

where P1, P2 and L are imported from a data file.

Answer 1

[Andrew Newell]

You can make a mesh:

[P1,P2] = meshgrid(P1,P2);

Then use interp2.

EDIT: Corrected typo.

[Lauren E]

If I do this, I get a blank plot.

[Andrew Newell]

What dimensions are your arrays?

[Lauren E]

>> size(P1)

ans =

    99     1

>> size(P2)

ans =

    91     1

>> size(L)

ans =

    91    99

[Andrew Newell]

Sorry, I was careless with the notation. Here is a working example of what I mean, starting with an example from the MATLAB documentation:

x = -8:1:8; y = x;
[X,Y] = meshgrid(x,y);
R = sqrt(X.^2 + Y.^2) + eps;
Z = sin(R)./R;

If we ignore how the data were generated, we now have vectors x, y and a matrix Z, and we could plot

surf(x,y,Z)

Now lets' interpolate to a finer grid. I like the spline method because it makes a smoother interpolation:

Xq = -8:.5:8; Yq = Xq;
[Xq,Yq] = meshgrid(Xq,Yq);
Zq = interp2(X,Y,Z,Xq,Yq,'spline');
figure
surf(Xq,Yq,Zq)

[imran saeed]

Based on above discussion I made this funciton, it will make coding easier, you can call this function many times to make your function smoother and smoother,

function [X,Y,Z]=smooth_spline(x,y,z)

n=length(x); sp=min(x); ep=max(x);

X=linspace(sp,ep,2*n);

sp=min(y); ep=max(y);

Y=linspace(sp,ep,2*n);

[Kxx,Kyy]=meshgrid(X,Y);

Z = interp2(x,y,z,Kxx,Kyy,'spline');

end