Fit x,y values to 2-D measured data set?

I have a small (3x11) set of 2-D high-quality measured data points from a scientific experiment: z(x,y). These define a fairly smooth surface. While it is easy to find z given any values of x,y by interpolation, I need to find all valid values of x,y in the domain for a given z. Please can anyone tell me if there is an easy way to do this in MATLAB? I guess least-squares minimisation fitting would be the best technique?

Answer 1

[Bjorn Gustavsson]

Well, I'm reaching a bit, but shouldnt this be something like the controur-lines at level z. If so then you have the tricont/tricontour functions if you want to do it in one go, or you could use first an interpolation functions: griddata, TriScatteredInterp and for example gridfit on the file exchange followed by a regular contour.

[Paul]

Hi, Bjorn

Thanks very much for pointing me to tricontour() - just what I needed