sarproz-group

Life time index for point selection

Tagged: 

This topic contains 3 replies, has 2 voices, and was last updated by  periz 4 months, 2 weeks ago.

  • Author
    Posts
  • #2049

    mahtab
    Participant

    Hi Dr.Perissin
    I want to select the Sparse Points based on their life time index, to use in the QPS technique. But when i choose the life time from parameters and press ‘Go’, no Sparse points are selected.
    What should i do ?! in other words, Which parameter should be used for Points selection in the QPS technique ?!

    Thanks.
    Mahtab

  • #2050

    periz
    Keymaster

    Hi Mahtab,
    the life time is calculated from amplitude time series of good scatterers using this module http://sarproz.com/manual/amp_series_proc.html. The analysis is carried out on the amplitude series of single pixels.
    The easiest way is to use the on-off model, then Tin and Tfin (birth and death dates) of targets will be estimated, their difference in number of images is the life time.
    This parameter is useful for good scatterers (you have to initially select a set of points, you can use the reflectivity map as thresholding parameter to get good bright scatterers).
    This parameter is not suitable at all for QPS.
    QPS are extended targets that can be more or less coherent because of many reasons: geometric decorrelation, seasonal or even random variations and so on. They are not simply temporary and you do not recognize their coherence variations using the amplitude of a single pixel.
    For selecting QPS candidates you can use the spatial coherence, calculated using the set of interferograms you want to use in your QPS analysis.

  • #2051

    mahtab
    Participant

    Hello again

    Thank You Doctor. I got it. But the Number of selected Points with spatial coherency index is less than ones with Amplitude stab. index.
    In my case, most of the points have low spatial coherency.

    Result of selected Points using spatial coherency and amplitude stab. indices with the same threshold are attached here.

    Attachments:
    You must be logged in to view attached files.
  • #2053

    periz
    Keymaster

    There are many things you don’t say,
    as how you calculated those parameters, what does your dataset look like (in terms of number of images, distribution of acquisition dates/baselines/doppler). then we should check the original parameters as reflectivity, ampl stab ind, spat coher and so on.
    anyway, as a general answer, if the spatial coherence is low and the ampl stab index is high, then this is not a case to process with QPS…

You must be logged in to reply to this topic.