python - ndimage map_coordinates with masked arrays -

python - ndimage map_coordinates with masked arrays -

i using ndimage interpolate follows:

ndimage.map_coordinates(input_data, coords, output, 2, prefilter=false)

now, problem not have valid measurements on whole input data. so, have masked array tells me info points valid. so, when doing interpolation, utilize pixels valid , adjust weightings accordingly (to ensure weights sum 1).

however, see there no easy way this. wondering if knows of way or can point me library or code can use. coming c++ background, still finding way around python.

it sounds need focus on interpolation of info , extract values desired coordinates. 1d splrep , 2d bisplrep interpolation functions need check out (a overview). both of these functions can weighted , provide fine tune command on spline function interpolate with.

once have filtered info desired weights can determine value @ specified coordinates using.

ndimage.map_coordinates(input_data, coords, output, prefilter=true)

note prefilter key word argument not needed default value

python scipy ndimage