# Plot contours using discrete data points

67 views (last 30 days)
mukesh bisht on 5 Jun 2021
Commented: Star Strider on 6 Jun 2021
Hi. I have three matrices (X, Y & Z) with X & Y contains x, y coordinates and Z contains value at this location. I plotted a coloured plot using "scatter function" as shown below. But I want to plot the similar diagram as contour plot not like colouring discrete points based on values. Please help SALAH ALRABEEI on 5 Jun 2021
% try this
contour(X,Y,Z)
or
% this
contourf(X,Y,Z)
mukesh bisht on 5 Jun 2021
No it is not a function. Corresponding to each point P(x,y), matrix Z contain a single discrete value

Star Strider on 5 Jun 2021
Let say i have 100 points so my matrices(X,Y & Z) are 100 x1
They could be gridded already. If they are, then simply using reshape the same way on eadh vector will create the necessary matrices. The easiest way to determine that is to plot them as:
figure
stem3(X, Y, Z)
grid on
If they are gridded, the regularity will be immediately apparent.
Otherwise, do something like this —
xv = linspace(min(X), max(X));
yv = linspace(min(Y), max(Y));
[Xm,Ym] = ndgrid(xv, yv);
Zm = griddata(X, Y, Z, Xm, Ym);
figure
contourf(Xm, Ym, Zm)
axis('equal')
Providing that there are no serious prolbmes with the origina vectors, that should work.
.
Star Strider on 6 Jun 2021
Having the data to experiment with made all the difference!
To make the contourf plot appear smoother, increase the number of contours. (I increased them to 200 here. Increase them further as desired.) The ‘steps’ are still there, there are simply more of them.
X = T1{:,1};
Y = T1{:,2};
Z = T1{:,3};
xv = linspace(min(X), max(X));
yv = linspace(min(Y), max(Y));
[Xm,Ym] = ndgrid(xv, yv);
Zm = griddata(X, Y, Z, Xm, Ym);
figure
contourf(Xm, Ym, Zm, 200, 'LineStyle','none')
colormap(turbo)
colorbar
axis('equal') Rotating it and putting it in the circle would be possible, however I would need to know how you arrived at that.
.