S1 GRD product_pre-processing for classification

Hello,

for the classification, is it neccesary to perform first Termal noise removal? In the help menu, it is defined as “Thermal noise correction can be applied to (…) Level-1 GRD products which have not already been corrected.” I’ve just want to know Your opinion about that (pro/cons).

And second question would be about multilooking, the description of the GRD product at ESA website is “(…) products consist of focused SAR data that has been detected, multi-looked and projected to ground range using an Earth ellipsoid model.” And still in some research and tutorials, authors are performing multilooking? For classification, is it enough to perform just speckle noise filtering?

Thank You very much in advance