e-sensing / sits

Satellite image time series in R
https://e-sensing.github.io/sitsbook/
GNU General Public License v2.0
440 stars 76 forks source link

Standardize the use of ROI in data cube regularization #1151

Open EstefaPizarro opened 1 month ago

EstefaPizarro commented 1 month ago

Describe the requested improvement When regularizing a data cube for a bounded region, smaller than a full tile, different satellites handle geometry cropping differently. While Sentinel-2 generates a regular crop by taking the maximum dimensions of the vector file, Sentinel-1 and COP-DEM mask the geometry. This generates disagreements in a possible subsequent treatment of a "merge" cube, because, although the union of the different cubes generated is allowed, the bands that make up it do not have the same dimensions of rows and columns. For example, it is not allowed as input for the segmentation process.

OldLipe commented 2 weeks ago

Create an internal function called .cube_crop() that will be used in sits_mosaic() and sits_regularize().