[Insight-users] Suggestion to a more flexible Interpolation function?

motes motes mort.motes at gmail.com
Sat Aug 29 17:21:38 EDT 2009


I am currently working on making a more flexible interpolation
function than the one used in:

itkBSplineInterpolationWeightFunction.txx

I would like to make it work for any points specified in physical
coordinates and without using the grid indexes. My idea is to
implement a BSpline function that computes the weights as a function
of the distance to k-nearest points from the source.

As an example look at the below image:

http://img42.imageshack.us/img42/2237/colorinterpolation2.png

Assume that I would like to compute the color value of the point W. I
know the values of the R,B,Y and G but I need to compute the weight
for each of these values to determine the value of W.

The unknow weights are: \alpha_R, \alpha_B, \alpha_Yand \alpha_G. I am
thinking on using the distance to the points as a parameter. I assume
that this is a pretty basic interpolation problem but I could use some
hints on getting started.


More information about the Insight-users mailing list