Hello,
i always see multilook applied after calibration so now i’m wondering if this is mandatory or it can also be applied before.
Can anyone explain what’s there right procedure and if possible also why it is?
Thanks in advance.
Calibration (in its basic form) applies a constant to each pixel to convert it into a comparable unit. This constant can vary between different acquisitions and takes into account the sensor conditions at the time and position of the measurement.
Multi-looking changes pixels values, mostly by aggregation in range direction. It makes therefore more sense to me to apply this specific constant to the original values.
The some applies to speckle filtering. Yet, none of them is mandatory.
so do you mean that i have to apply multilook after calibrations? Sorry for repetition.
No, afterwards. Calibration (so I think) should be the first step.
Ok, and again thanks for your help!
I was just wondering if mathematically speaking applying a constant to some values or to their mean (is it mean applied in multilook?) is equivalent.
Well, if extreme values were corrected during tthe calibration, for example, it could make a difference, also when the incidence angle is integrated in the calibration.
It is rather a feeling of mine, maybe someone can comment.
For sure your opinion is based on very solid fundations.
I take the chance to ask you another thing, somewhere in the forum i read that you are used to apply multilook on GRD products as a way to resample the image, is it correct?
yes, GRD products are already of square pixel size (10 x 10) by definition. But if you investigate a large area and want to reduce the data size, you can still apply the multi-looking to reduce it to 20, 30 or whatever resolution. As a nice side effect, speckle is reduced.
Thank you very much.