Dear colleagues
To perform the supervised classification of an S2 image in SNAP, I have defined several training areas as vectors using the SNAP functionalities (“Add a new vector container”) in the original image. SNAP creates the corresponding masks (geometry_1, geometry_2, etc.). Them, I’ve created a subset of the image, and both the vectors and the corresponding masks are available in the subset (no problem at this stage). Then, to use all the bands in the classification, I need to perform the resampling of the subset image, so all the bands have the same spatial resolution (otherwise I get an error message). However, in the reprojected product, the vectors and the masks corresponding to the training areas no longer exist, i.e., they are not saved in the reprojected product. I have tried this procedure several times and with different S2 images and the result is the same.
I have done this several times with version 7 and never had this problem.
What am I missing? Does someone have the same problem?
Thanks in advance,
Ana