Hi Jaap,
could you expand on what you mean with filter entire images based on the result of a calculation ?
From what I understand, it could be something like:
- Computing the average over the datacube, reducing the spatial dimensions (the image would be the data of each timestep in the datacube)
- If the average is above a threshold keep it otherwise discard it.
Basically, something similar to what a back-end is doing for filtering out the dates with too many clouds Filter less cloudy image - #5 by michele.claus
I will see if it is currently possible with the available processes, maybe reduce_spatial could be replaced with two reduce_dimension in a row, firstly over x and secondly over y or viceversa.