how to display a vector as colorplot of to other vectors

Hi, I couldn't figure that one out myself, so hopefully somebody has a solution.

I have three vectors X, Y, and Z of same dimension, so that Z(k) is a height for horizontal coordinates (X(k),Y(k)). I want to display Z as a color plot over the horizontal area that encompasses all coordinate pairs of X and Y. All functions I know (pcolor, surf, etc) require Z to be a matrix which is not really feasible due to memory problems since the vectors are quite large (appr. 250000 elements). Also, the horizontal coordinates are not on a regular grid but more like a point cloud.

Any idea how to display the data as it is would be appreciated.

Cheers, Peter

[Vaibhav Awale]

Hi,

Can you please explain how Z(k) is the height for horizontal coordinates (X(k),Y(k))? I mean suppose Z(1) is height for (X(1),Y(1)), but then what is the height at (X(1),Y(2))? Is it still Z(1) or is it Z(2)?

Regards, Vaibhav

[Image Analyst]

It's just a list of 3D points:

x1, y1, z1
x2, y2, z2
x3, y3, z3
etc.
xn, yn, zn

Each row is one point, though he has it in 3 different arrays: x, y, and z. You would never pair things from different rows - it doesn't make sense. So if you're looking at the 4th point, you would use x(4), y(4) to get the lateral location, and then z(4) to get the Z height. You would never use x(2) and y(37) because there is no point there. Does that make sense now?

[Vaibhav Awale]

Yes, thank you for explaining that. scatter3 or plot3 will work well in that case as suggested by Star Strider.

Another option might be to map values in z vector to RGB values and make a plot of each point with the corresponding color from z vector.

Answer 1

[Star Strider]

If you have a point cloud, the scatter3 or plot3 functions would work. There are a number of options for coloring the points.

If you want to grid them at some point, see if scatteredInterpolant will do what you want.

[Image Analyst]

I was also going to suggest the same thing: first get a regular image with scatteredInterpolant(), then display with some normal display function like imshow(), surf(), or whatever. Some other cool visualization options are in the MATLAB Gallery: http://www.mathworks.com/products/matlab/plot-gallery.html