CODEWITHSUNDEEP
matlab
user3008019
user3008019

Reputation: 11

plotting surface of a matrix in matlab

I have a matrix t with three columns x,y and z that contain radius, angle and signal values from my simulation, respectively:

t = [
x   y    z
10  0   -1
10  2   -11
10  3   -10
20  5   -25
20  7   -100
20  20   -101
20  3   -12
30  12   -104
.    .     .
.    .     .
.    .     .
]

as you can see, the values have not any order, and also the first column (x) has a lot of repeated values. What I want is to make a nice surface that shows the fluctuations of the signal values (z column) in respect to the radius and angle values (x and y columns). I searched a lot and I tried almost every suggestions in the site, but always faced with errors like data dimensions must agree. Can someone help? thanks in advance.

Upvotes: 1

Views: 1213

Answers (2)

Dennis Jaheruddin
Dennis Jaheruddin

Reputation: 21561

I believe this is what you need:

M = [10, 0, -1
10, 2, -11
10, 3, -10
20, 5 , -25
20, 7 , -100
20, 20, -101
20, 3, -12
30, 12, -104];


[xq,yq] = meshgrid(min(M(:,1)):1:max(M(:,1)), min(M(:,2)):1:max(M(:,2)));
vq = griddata(M(:,1),M(:,2),M(:,3),xq,yq);

surf(xq,yq,vq)

In order to plot a surface you need to decide what the value is between the used points. In this case I have used griddata to interpolate.

Upvotes: 0

3lectrologos
3lectrologos

Reputation: 9662

You can use your available points to get interpolated points on a regular grid and then create a surface plot as described here:

xs = linspace(min(t(:,1)), max(t(:,1)), 50);
ys = linspace(min(t(:,2)), max(t(:,2)), 50);
[x, y] = meshgrid(xs, ys);
f = scatteredInterpolant(t(:,1), t(:,2), t(:,3));
z = f(x, y);
surf(x, y, z);

Another way would be to triangulate your existing points and use trisurf:

tri = delaunay(t(:,1), t(:,2));
trisurf(tri, t(:,1), t(:,2), t(:,3));

Upvotes: 1

Related Questions