X
PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 04 Jan 2015 12:05 PM by  anon
Interferometric workflow, master file, slave file
 5 Replies
Sort:
You are not authorized to post a reply.
Author Messages

anon



New Member


Posts:
New Member


--
04 Jan 2015 12:05 PM
    This regards to input files in Interferometric workflow: What happens if I put as a master file the older (3.5.2012) TerraSar x .slc file, and as a slave file newer .slc file (14.5.2012)? Is it mandatory that master file is "older" than slave file? What is master file and what is slave file? I ran the process with two TerraSAR x files (master file=newer date, slave file=older date) and it couldn't finish... Thank You very much. Filip

    Deleted User



    New Member


    Posts:
    New Member


    --
    05 Jan 2015 02:13 PM
    Now I read somewhere that: It does not matter which of the input data are selected as master and which is slave, it is mandatory that the initial order is maintained throughout the entire processing chain. But my phase unwrapping process is not working well or is very Very time consuming (Minimum Cost Flow and Delaunay MCF). It stops around 8 % for like 5 hours or more. I am using this unwrapping for entire TerraSar x scene. I dont know what is the problem...

    Deleted User



    Basic Member


    Posts:228
    Basic Member


    --
    06 Jan 2015 01:01 PM
    It sounds like you may want to contact Technical Support or you local distributor to track down what is going wrong in your case. It is possible that the processing is still active, and it just takes a long time. Can you see the process continuing to take system resources? In that case, it is still working. Giving it more time may allow it to complete. Peg Exelis VIS

    Deleted User



    New Member


    Posts:
    New Member


    --
    07 Jan 2015 04:09 AM
    Yes, the process was active for too long and then I quit the processing. However, phase unwrapping with region growing algorithm was completed successfully unlike MCF.

    Deleted User



    Basic Member


    Posts:228
    Basic Member


    --
    07 Jan 2015 11:43 AM
    This could be normal and expected. The MCF algorithm is very slow and requires lots of memory. The Region Growing option is much faster. Normally you would only use the MCF algorithm if the output from using Region Growing looks bad and/or there are large areas with low coherence. - Peg Exelis VIS

    Deleted User



    New Member


    Posts:
    New Member


    --
    11 Jan 2015 02:58 AM
    Thanks for your help. Yes, the outcome with region grow looks very bad. Digital Elevation Model is not good and MCF is very slow. I will try to subset my files and hopefully that will speed up the process.
    You are not authorized to post a reply.