Scribd
Upload
7 views

Notes 3

Uploaded by

geologiteknikpatgtl
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views

Notes 3

Uploaded by

geologiteknikpatgtl
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

APS Processing

To start the APS processing, press the Go button next to the APS estimation label in the Multi-Image
InSAR processing group.

The APS Processing window opens. There are quite a lot of options to choose from. We are going to
address the standard parameters step-by-step. In the beginning, we need to select the sparse points
used for the processing. For this, we have to select a parameter used for the sparse point selection.
There are quite a lot of parameters available. A good standard choice is the Amplitude Stability Index 1
– Sigma/Mu. Therefore, we are selecting Amp. Stab. Index 1 – Sigma/Mu in the Parameters box of the
Sparse Points Selection group. As threshold, we enter 0.8 in the Thresh. text box. Then we press Go in
the Sparse Points Selection group. Afterwards you can plot the selected points via the Plot button in the
Sparse Points Selection group.
Now, we connect the graph using the Go button in the Graph Creation group. With the standard
parameters a Delauny triangulation is used, which should be good in most cases. Afterwards you can
plot the selected points via the Plot button in the Graph Creation group.

First, we chose some standard parameters for the processing. We estimate the linear trend between -
100 to 100 mm/year and the height between -100 to +100m around the SRTM height. We therefore use
the Ext. DEM for the height estimation.
Afterwards, we press Go in the Connections processing group. This may take a while. After finishing , we
check the Histogram and the connections Graph in the Connections coherence group by pressing Hist
and Plot Graph.

They look good. We keep the standard parameters for the Non-Linear Weighing and press Plot in the
Non-Linear Weighing group.
Looks OK, so we can continue. We calculate the reference point by pressing Go in the Reference Point
group. Then we estimate the parameters by pressing Plot in the Estimated Parameters group.
No the difficult part starts. We have to analyze the results and decide if they are OK and we continue, or
if we have to adjust some parameters. This is a matter of experience and also requires a willingness to
play with the parameters to see their effect. Well, in the results above, the heights look OK, at least we
can see some of the high buildings are estimated correctly. However, the velocity is quite noisy and we
are not satisfied.

One parameter to adjust easily, without needing a long reprocessing, is the non-linear weighing. We are
getting a bit more aggressive and going for 0.7, 0.85, 1
Again we recalculate the reference point and plot the estimated parameters.

Well, this looks already much better. The histogram of the integrated residual heights is shifted to the
left, which might be because the selected reference point is on a building and not on the ground. We
can select another reference point by pressing Load in the Reference Point group.
SARProZ is automatically creating five points called AutoRef1 to AutoRef5. We select AutoRef3 and plot
the estimated parameters.

The height shift has gone. The results look OK and we could continue processing. However, we see that
in our height estimation, the value of -100 to +100 might not be an optimal choice, because there are
quite high buildings in the Las Vegas scene.

We are therefore changing our processing parameters. We are trying to estimate the height from -50 to
150m. We also saw in our first results that the velocity might be not so high. Therefore, we reduce the
estimation range from -60 to 60 mm.
Again we press Go in the Connections processing group to start the process and see the coherence by
pressing Hist and Plot Graph in the Connections coherence group.

We are still using 0.7, 0.85, 1 for the non-linear weighing.


We again select a reference point by pressing Go in the Reference Point group.

Well, this time we are satisfied with the results. However, we could (and probably should) continue to
search for optimal parameters in the above described way.
Finally, we go for the APS processing selecting the right option in APS options. I.P. stands for inverse
parameters and is a rather stable estimator for the APS. I.R. stands for inverse residuals and can offer
better results in good cases, which means cases with coherent and a large amount of data. Tri. I.R. is
using triangulation for the inverse residuals. Stratif. stands for stratified and can be used when different
atmospheres in different heights are to be expected. Be aware, that to use this successfully, a larger
amount of images is typically necessary.

In our example, we select I.R. because we have a very stable and coherent dataset and we are
optimistic. The APS is estimated by pressing Go in the APS Estimate group. Afterwards we can plot
through the estimated APS by pressing the Plot button in the APS Estimate group.
By pressing the Test button in the APS Estimate group, we can get an estimation of the APS estimation
quality.

Finally, we can close the APS estimation by pressing the OK button.

You might also like