I'm working with the Cross Ratio (VH/VV) derived from Sentinel-1. I'm trying to monitor vegetation and I'm looking only at the spatial aspect of the signal.
I've read quite a lot of literature on the topic, and I see the spatial filters (to reduce speckle) are always applied on the separate polarizations before computing the Cross Ratio. However, when I compare my results with NDVI, I get a better spatial correlation when I apply the filters after. I've tried many different filter settings, so it cannot be that the configuration I use coincidently matches better with the Cross Ratio.
Does anyone have experience with this? Happy to discuss further!