sarproz-group

Radarsat-2 data import

This topic contains 12 replies, has 3 voices, and was last updated by  espiritocz 2 years, 3 months ago.

  • Author
    Posts
  • #771

    cnu_lee
    Participant

    Hi,professor

    How to import radarsat-2 data?

    My soure image is TIFF format,it’s in a folder called”RS2_OK60115_PK140099_DK124064_W1_20101122_222717_VV_SLC”,its original name is image_VV.tif,in these folders ,there are other XML format file.
    I try to put one TIFF format in SLC folder,the work directory is like this “/home/user/rsat-2/”.It didn’t work,SARProz couldn’t identify my image.
    So I tried another means,I put the whole “RS2_OK60115_PK140099_DK124064_W1_20101122_222717_VV_SLC” folder in SLC folder,and set the working directory.But it didn’t work either.
    I don’t know how to solve this problem.

    Thanks you,professor.I hope to get your response when you are not busy.

  • #772

    periz
    Keymaster

    Based on the Radarsat-2 data we processed up to now,
    the software searches for directories containing the string “_RST2”.
    I suggest you at this moment to manually rename your directories including such string.
    We may change the code later when we will confirm the naming convention.

  • #785

    espiritocz
    Participant

    Hi,
    same problem here – I tried to solve it by renaming all RS2 directories to contain _RST2 string but it didn’t help:

    First identified directory: /C/Process_SARPROZ/Brno_radarsat/SLC/RS2_OK55634_PK527325_DK467628_F23_20140808_163838_VV_SLC_RST2

    get_dir_slc: no known format recognized in directory /C/Process_SARPROZ/Brno_radarsat/SLC/

    thank you.

    Milan

  • #786

    periz
    Keymaster

    sorry, the string _RST2 must be included but the directory must also terminate with _SLC
    if you rename it e.g. RS2_OK55634_PK527325_DK467628_F23_20140808_163838_VV_RST2_SLC it will work

  • #787

    espiritocz
    Participant

    Thank you, however it still doesn’t work:

    First identified directory: /C/Process_SARPROZ/Brno_radarsat/SLC/RS2_OK55634_PK527325_DK467628_F23_20140808_163838_VV_SLC_RST2_SLC

    get_dir_slc: no known format recognized in directory /C/Process_SARPROZ/Brno_radarsat/SLC/

  • #788

    periz
    Keymaster

    right, this works:
    RS2_OK55634_PK527325_DK467628_F23_20140808_163838_RST2_VV_SLC

  • #789

    espiritocz
    Participant

    yes, it works now.

    thank you

  • #790

    espiritocz
    Participant

    hi,
    i get another issue – it seems that Radarsat-2 gets wrong rotation – the GCP is set properly (manual match of airport object), but the whole image is shifted to the South while it should be to the North. Thus also DEM doesn’t match.
    It seems to me that the software expects different mirroring/orientation of the image. Maybe a different format?
    I have ascending fine mode VV data.
    Sending screenshot – the “red” extent is original kml from CSA – the extent below is the mirrored amplitude mean from Sarproz – you can see in the overlapping area the airport that matches well.

    Thank you for comments/update.

    Milan

    Attachments:
    You must be logged in to view attached files.
  • #792

    cnu_lee
    Participant

    Hi,professor,

    I get two parts of RST-2 images, the one is .rlsc format co-registered by GAMMA with .par file, the other one is raw data, (e.g. RS2_OK60115_PK140099_DK124064_W1_20101122_222717_RST2_VV_SLC). But the two parts of images aren’t during the same period.I hope to put them together to process.Is it OK in principle?

    I create a SLC folder and RSLC folder to put these data, but SARPROZ couldn’t recognize the RSLC folder, even though the raw data could be recognized in SLC folder.

    Could the process work in another import way?Or it can’t be realized at present?

    Thanks for your comments,sir.

  • #793

    periz
    Keymaster

    CNU_lee: at this moment you cannot mix gamma RSLC images with new SLC images… in principle it’s possible but at this moment the sw is not considering such option… we will implement it in the future, but for now I suggest you to find a different solution…

    Milan: I see such thing for the first time. Probably the orbital data are flipped. Can you share one image with me for debug?
    d

  • #794

    espiritocz
    Participant

    Hi, thank you.
    indeed, the orbits have huge errors, showing also strong phase ramps.
    I will send you one image for reference using ftp (in private message). thank you!

    Milan

  • #801

    periz
    Keymaster

    Milan:
    the data were stored in a flipped format. Now the data extractor has been generalized and the image has the right orientation.
    I saw the big orbital data errors. However, we did not test the phase flattening yet.
    Let me know if you still get big ramps. If so, we’ll also generalize an orbit correction.
    Pcodes and compiled version for windows are now updated.
    The linux compiled version not yet: we’ll update it in the next days
    d

    • #803

      espiritocz
      Participant

      Dear Daniele,

      thank you for fast response!
      I will test it now (Manage Version shows the last to be from 27-Apr, is it the new one?).
      The topic of AOI is subsidence of buildings – three weeks ago a dangerous movement was (accidentally) detected over one building. With this dataset we may detect this movement in similar way you did for Hong-Kong, we’ll see.

      Milan

You must be logged in to reply to this topic.