umr-lops / xsar

Synthetic Aperture Radar (SAR) Level-1 GRD python mapper for efficient xarray/dask based processing
https://cyclobs.ifremer.fr/static/sarwing_datarmor/xsar/
MIT License
25 stars 8 forks source link

revisit of rasterization from annotations #57

Closed agrouaze closed 2 years ago

agrouaze commented 2 years ago

address issue #56 .

agrouaze commented 2 years ago

last commit allow to get interpolated variables from geoloc annotations in NESZ fashion. for now it is only in s1meta (not added in the sentinel1_dataset.py. The way to test is the following:

atrack = np.arange(10,900)
xtrack = np.arange(40,700)
values_hr = s1ds.s1meta.get_bursts_polygons(atrack,xtrack,s1ds.s1meta.geoloc['incidence'])
valuess = values_hr(atrack,xtrack,s1ds.s1meta.burst_azitime(atrack).astype(float))

this first draft implementation is quite slow on the whole subswath. It needs optimization.

agrouaze commented 2 years ago

24 March 2022: @oarcher and @agrouaze decided to

  1. ditch the rasterization of geolocation fields in the "NESZ" fashion
  2. compare the content of s1meta.gcps and s1meta.geoloc to be sure that we bring new informations with geoloc
  3. in a second time remove the methods related to gcps
agrouaze commented 2 years ago

image validation of a geolocation field from TOPS SLC IW over a change of bursts

agrouaze commented 2 years ago

comparison of s1meta.geoloc versus s1meta.gcps image image geoloc is bringing new fields and keeps informations previsouly provided by gcps.

agrouaze commented 2 years ago

commit https://github.com/umr-lops/xsar/pull/57/commits/36dca70c581da3555754656f6533ca00456f11ad close #19

agrouaze commented 2 years ago

latest dataset return : image

oarcher commented 2 years ago

Merged. Huge contribution to xsar @agrouaze ! :champagne: