splitting a dataset

Viewing 2 reply threads
  • Author
    • #1972

      Hello Daniele,

      (thank you for the implementation of the SRTM download change.)

      I am facing a special problem: to split an already coregistered dataset into two, resp. excluding a significant number of images (I need to perform the processing of installed reflectors separately after the data they were installed).

      But, in the dataset, the master is older. Is it possible to exclude the master from an already-coregistered dataset? Or, is it possible to perform some tasks (refl. map & amp stab computation, probably also small area processing) without some images (masking images)?

      It would be nice to have both datasets (before and after the installation) coregistered to a common master – and, as you understand, I do not want to do it all again with a newer master.

      (by the way: I expected that it is possible to have only new images in the SLC folder (and delete all those already coregistered. But that is not right, the master must also be there – if not, I get errors like “index exceeds matrix dimension”, and the number of images to update is not correct.)

      Thank you for any idea,

    • #1973

      Hi Ivana,
      there are several ways to do what you want to do.
      1. if you want to process temporary targets, you can analyze the amplitude time series with an on/off model and then use the amplitude as a weight in the insar time series processing frames
      2. you can select the images to process in the select dataset module limiting the range of temporal baselines (values of btin and btfin there are in days)
      3. you can act on the DataSet.txt file in the DATASET_INFO folder and keep only the images you want

      There should be no issue if the master is not included in the selection, you will simply get a message and the sw will take an other one as a reference. You can also choose the image you want as master in the dataset selection window

      but yes, this is something we will fix, at the moment the software needs the SLC file of the master if you want to update the dataset with new images…

    • #1974

      Hi Daniele,
      thank you.
      I tried points 2 and 3. With point 2, however, I could not see a difference in the reflmap (it seems to me that it is computed from all the images), but the problem was solved with point 3, with no problem (just warnings) with the master image.

      Regards, Ivana

Viewing 2 reply threads
  • You must be logged in to reply to this topic.