[math] BicubicSplineInterpolator - how to apply real-world data

[Andreas Mucha <si...@gmx.net>]

Hello all,
 
I have a question regarding BicubicSplineInterpolator .
I have to visualize measured data using interpolation.
 
The BicubicSplineInterpolator documentation says :
xval - All the x-coordinates of the interpolation points, sorted in increasing order.
yval - All the y-coordinates of the interpolation points, sorted in increasing order.
fval - The values of the interpolation points on all the grid knots: fval[i][j] = f(xval[i], yval[j]).
 
I have about 20 probes of x,y,z tuples. x,y is the location. z is the measured value.
If I would order the tuples by x, I have no order in y and vice versa.
Ordering x without y (and vice versa) would destroy the correct z assignment.
 
How do I feed the interpolator with my data ?
 
Many thanks in advance .
 
Andreas

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org

Re: [math] BicubicSplineInterpolator - how to apply real-world data

[Gilles <gi...@harfang.homelinux.org>]

Hello.

>
> I have a question regarding BicubicSplineInterpolator .
> I have to visualize measured data using interpolation.
>
> The BicubicSplineInterpolator documentation says :
> xval - All the x-coordinates of the interpolation points, sorted in
> increasing order.
> yval - All the y-coordinates of the interpolation points, sorted in
> increasing order.
> fval - The values of the interpolation points on all the grid knots:
> fval[i][j] = f(xval[i], yval[j]).
>
> I have about 20 probes of x,y,z tuples. x,y is the location. z is the
> measured value.
> If I would order the tuples by x, I have no order in y and vice 
> versa.
> Ordering x without y (and vice versa) would destroy the correct z 
> assignment.
>
> How do I feed the interpolator with my data ?

This interpolator requires that the samples are taken on a
regular grid; that's what the documentation tries to indicate.
If you have a random cloud of points, you could try the
"MicrosphereInterpolator" for which there is no such requirement.

Best regards,
Gilles


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org