This the graph for ship detection with the default settings.
The graph for preprocessing of the image with the default settings.
sorry, I meant as an xml file.
are you using the latest version of SNAP? The Land/Sea Mask operator struggles with the retrieval of SRTM 3Sec data, but this has been fixed in a recent update.
Not sure about the version as I am using a friend system. But the toolbox plugins are updated.
please open menu > Help > About SNAP > S1TBX
You find the exact version at the bottom:
The version is the same as the one you had messaged.
do you first apply the process.xml to the data and afterwards the detection.xml?
Because calibration is done twice then.
I first process the data and then go for the detection part. The Calibration is done twice thats correct and the same is case if I do all the steps manually myself but in that case I get the file for ship detection.csv but not when I do it with help of the graph.
Thank you for the response I was able to get the ship_detection.csv file now but when I run the Ellipsoid correction on it to get it geometrical corrected the result is different from before.
Pink - New result
Yellow - Old result
which of the ones comes closer to the reprojected image?
Could you please guide me to how exactly can I know that?
Export the geocoded raster image as kmz and display it in Google Earth as well. This will show you which points are more correct.
Getting an Error on trying to export it as kmz.
Please zoom in a bit more (reduces the size) and disable the detection results in the layer manager.
Please also make sure the data is projected to WGS84
What is the size of the scene?
KMZ can only handle images up to a certain size. Could be that your is to big.
And creating tiles of the image is not yet implemented:
You can try to export your image as Geotiff, and then import it into Google Earth. Then Google earth does the tiling, if necessary
It worked. Thank you!
How can I check whether the ship detection results are accurate with the SAR data?
Got no clue on how to proceed with it, any help is appreciated.