Hello! My question is about S1 SLC /GRD products.
Witch product is more useful for Monitoring of crop conditions, soil properties and mapping tillage activities?
Can I do some kind of crop classification and witch product is more useful for crop classification?
It depends on your classification algorithm. If you want to apply
polarimetric classification then you will need polarimetric SLC data
but, with dual pol its quite limited. If you classification is just
based on the backscatter then go with the GRD products.
Hi I have some question about change detection first of all what is the best data format for change detection? slc or grd / IW or EW or WV or SM/ and what polarization? second is there any tutorial or step by step processing guide for doing change detection?
First of all the mode most depends on the acquisition plan. IW is available over most land masses. EW is available at the poles or coasts. SM is only for emergencies or some volcanic islands.
The choice of SLC vs GRD will depend on your application. With SLC you can examine the complex coherence between the images. Google tutorials on ‘coherent change detection’. This can be good to see small changes between images that are very close together in time.
With GRDs you can do amplitude change detection and this can be good to see large difference over a longer time period for example to see how a city has grown across a decade.
The most important step in the processing is to precisely coregister the images so you know you are comparing the right pixels between the images. Then you may want to apply the radar/SAR applications/change detection operator with the log ratio. You then may want to use the color manipulation to threshold and colour code high positive and high negative changes with red and blue and values in between you could make transparent.
thanks for your detailed answer.
I want to do change detection in urban areas with sentinel 1 IW SLC data with 1 year difference in between, in order to what you said, do you think is it good for this purpose?
and if I want to do this in SNAP what do you offer step by step process ?
GRD may be easier. You could try calibration, coregistration and then amplitude coherence and amplitude change detection.
You could also try terrain mask and speckle divergence to make the urban areas stand out and then use this as a mask on your change detection.
Thank you so much Lveci
what is amplitude coherence? is it an option in SNAP? how can I do it?
unfortunately I can’t get what you mean
I want to do change detection on slc format , can you offer me a standard method on urban area in SNAP?
because I want to know what change appears in urban area!
I am so appreciated
SLC:s are useful if and only if you are going to do InSAR or polarimetry, otherwise working with GRD:s will be a lot faster and simpler…
1 Like
SLC are for insar or polarimetry
If you coregister GRD products, you may still run the coherence operator on it. It will give you a kind of amplitude coherence.
Thank you so much Mengdahl and Lveci
sorry to dig this out but is there a theoretical background behind this “amplitude coherence”?
It’s simply a measure of the similarity of the intensity speckle patterns. It should be better than nothing but no substitute for complex InSAR coherence.
thank you. So is it just r² for the given moving window?
1 Like