Appendix D
Regridding versus resampling

Sometimes you may wish to put your data onto a new pixel-size scheme. This may either involve regridding (often called rebinning) the data onto larger pixel sizes, or resampling the data onto smaller pixels, both using Kappa.

The key difference between rebinning and resampling is whether you iterate over the input or output pixels. Rebinning divides each input pixel value between a group of neighbouring output pixels, whereas resampling allocates each output pixel a value sampled from the input array.


pict
Figure D.1: Regridding versus resampling.


resampling—Resampling of the grid of input data is performed by transforming the co-ordinates of the centre of each output pixel into the co-ordinate system of the input grid. Since the resulting co-ordinates will not, in general, coincide with the centre of an input pixel, sub-pixel interpolation is performed between the neighbouring input pixels. This produces a resampled value, which is then assigned to the output pixel. Since resampling does not integrate, the total data value in the input image will not, in general, be conserved. Resampling has the advantage of ensuring that the output image is filled (provided the input array contains good data).

rebinning—Rebinning of the grid of input data is performed by transforming the co-ordinates of the centre of each input pixel into the co-ordinate system of the output grid. The input pixel value is then divided up and assigned to the output pixels in the neighbourhood of the central output co-ordinates. A choice of schemes are provided for determining how each input pixel value is divided up between the output pixels. In general, each output pixel may be assigned values from more than one input pixel. All contributions to a given output pixel are summed to produce the final output pixel value. Rebinning can leave gaps in the output array if the output array’s pixels are smaller than those of the input array.