Interp3. Error using griddedInterpolant. The grid vectors do not define a grid of points that match the given values.

Hi,

I'm trying to use interp3(x, y, z, image, xq, yq, zq).

My new query xq, yq and zq are [256, 256, 90], and my old x, y and z are equal to [196,256,512].

I get the following error

Error using griddedInterpolant
The grid vectors do not define a grid of points that match the
given values.

Error in interp3 (line 146)
            F = griddedInterpolant(X, Y, Z, V, method,extrap);

Error in clinicalTest_aligment (line 99)
newMrImg = interp3(X, Y, Z, oldImage, Xq, Yq, Zq);

It does not work because the x is smaller than xq?

In that case it is 'extrapolation' and not interpolation?

Thanks in advance

[Adam Danz]

Could you explain the bigger picture of your goal? The x,y,z values are the 1x3 vectors you mentioned is fishy and isn't gridded. Nevertheless, the error is not reproduced given the inputs you provided.

x =  [196,256,512]; 
y = x; 
z = x; 
v = rand(3,3,3);
xq = [256, 256, 90]; 
yq = xq; 
zq = xq; 
Vq = interp3(x,y,z,v,xq,yq,zq)  %no error

Perhaps using griddedInterpolant() would be a better approach but the question about the bigger picture of the goal is still relevant.

BTW, nice handle and avatar from a fellow drummer and brain reseacher.

[drummer]

Hi Adam, I am trying to interpolate a MRI volume image to a smaller size.

Previously I 'meshgridded' x, y and z and xq, yq and zq, so:

[x, y, z] = meshgrid(xDim, yDim, zDim)
[xq, yq, zq] = meshgrid(newxDim, newyDim, newzDim) % newxDim, newyDim and newzDim are 2-dimensional vectors
% to compose the new dimensions of Vq. Ex: newxDim = xVector(256,1) = newyDim, and newzDim = zVector(90,1)
% same for xDim, yDim and zDim.

So the dimension being smaller in the original image it is not related to the issue I'm facing?

You're actually the first fellow drummer neuroscientist I ever met. Cheers.

[Adam Danz]

I do extracellular recordings so I've spent a lot of time (4-6 hrs a day for years) listening to spike trains of visually responsive neurons (MSTd) firing in real time to various visual stimuli. I've always thought it would be cool to combine segments of those audible responses to dubstep music :D

Answer 1

[Adam Danz]

Follow this tutorial and if you get stuck or come across a problem that isn't already addressed in the forum I'd be glad to dig in. If you're working with an image of uint8 data, you'll need to convert it to double, first.

https://www.mathworks.com/help/matlab/math/resample-image-with-gridded-interpolation.html