This soubroutine interpolates the grid with a polyharmonic thin-plate spline. This subroutine needs the LUPACK and BLAS libraries to evoke the dgesv subroutine. The method uses the Thin Plate Spline. We use these resources: http://cseweb.ucsd.edu/~sjb/eccv_tps.pdf http://www.univie.ac.at/nuhag-php/bibtex/open_files/po94_M%20J%20D%20Powell%2003%2093.pdf http://vision.ucsd.edu/sites/default/files/fulltext(4).pdf
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| double precision, | intent(in) | :: | grid_coarse(n_coarse) |
Coarse input grid for interpolation. |
||
| double precision, | intent(in) | :: | gamma_coarse(n_coarse) |
Coordinates of on coarse grid. |
||
| double precision, | intent(in) | :: | pparbar_coarse(n_coarse) |
Coordinates of relativistic parallel momentum on coarse grid. |
||
| integer, | intent(in) | :: | n_coarse |
Number of entries in coarse grid. |
||
| double precision, | intent(in) | :: | gamma_rel(nspec_rel,0:ngamma,0:npparbar) |
Coordinates of on fine grid. |
||
| double precision, | intent(in) | :: | pparbar(nspec_rel,0:ngamma,0:npparbar) |
Coordinates of relativistic parallel momentum on fine grid. |
||
| integer, | intent(in) | :: | ngamma |
Number of steps on fine output grid. |
||
| integer, | intent(in) | :: | npparbar |
Number of parallel momentum steps on fine output grid. |
||
| double precision, | intent(in) | :: | smoothing |
Smoothing parameter for spline interpolation. |
||
| double precision, | intent(out) | :: | f0_rel(nspec_rel,0:ngamma,0:npparbar) |
Fine output grid after interpolation. |
||
| integer | :: | is_rel |
Index for relativistic species (if any). |
|||
| integer | :: | nspec_rel |
Number of relativistic species. |