surface fitting1
3 次查看(过去 30 天)
显示 更早的评论
Hi,
I'd like to ask for help regarding surface fitting and I have very little experience in Matlab.
I have a Z 77x77 matrix.
If I use surf(Z) I can plot it without problem.
I would like to interpolate Z but I couldn't get it right neither with the sftool nor with the gridfit.
If I call sftool I don't even have an option to select this Z matrix. I've read the manuals but it is still unclear how I could do this.
So the question is how to use this interactive sftool if I have a 77x77 large Z matrix [measurement data] and I want to interpolate or somehow fit it.
Thanks, Krivan
0 个评论
采纳的回答
John D'Errico
2011-6-3
Why use gridfit, or for that matter, the surface fitting tool? It is already a surface, on a regular grid. Use interp2, a tool designed for exactly that purpose.
help interp2
Edit: If x and xi are identical, then why are you trying to interpolate anyway? You would get out the input array even if you had called interp2 properly! It does not sound as if you are trying to do interpolation at all.
As far as the surface goes, if it is too noisy and you just want to make it less so, then do some smoothing. A simple solution there is to use a convolution tool, although then you need to deal with edge effects. Easier is to use gridfit, with a larger than normal value for the smoothness parameter.
[xmesh,ymesh] = meshgrid(x,y); zgrid = gridfit(xmesh(:),ymesh(:),z(:),x,y,'smoothness',k);
By default, the smoothing parameter is 1, a value which will yield relatively little smoothing. Try 2 or 5 or 10. Pick a value that makes you happy.
0 个评论
更多回答(3 个)
Rob Graessle
2011-6-3
To do surface fitting you need to define your X and Y data vectors.
X=1:77; Y=1:77; % Or any other vectors of 77 points
Then you can use SFTOOL by supplying X, Y and Z as the data vectors.
0 个评论
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Get Started with Curve Fitting Toolbox 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!