I am working with Sentinel-1 SAR GRD images in IW swath in HH polarization. I have a few questions to ask about preprocessing and also improvement in backscatter intensity to identify ice.
I preprocess the sentinel-1 image by thermalnoiseremoval>calibration>specklefilter>terraincorrection.
My final output is a Sigma0 band.
Before I starting with preprocessing, do I need to convert the Amplitude and Intensity to dB value or is it after I complete my preprocessing that I need to convert the Sigma0 to dB value.
There is lot of noise in the ocean pixels and it times it is difficult to differentiate between ice and water. Is there a way to improve the differentiation between the two? The way I am thinking is to implement non-uniform filters such as “median” followed by morphological filters such as “closing” to improve the edges.
I also tried to use gradient filter such as Sobel. However I am not sure how to implement the resultant image after sobel filteration. Can you throw some light on how to use it in combination with other filters?