test_noise.m: Apply SigmaZeroSFPoly to measured noise value #4
Conversation
There was a problem hiding this comment.
Radiometric.NoiseLevel.NoisePoly is an absolute measure, not a calibrated measure.
It would be OK to display measured noise in sigma-0 (and probably even preferable, since that's what most SAR people speak in), but we would need to apply the scale factor to both the predicted and measured numbers, not just one. RCSTool.m has an example of code displaying predicted noise in a calibrated way.
Also, it is possible for a SICD to have NoisePoly populated, but not SigmaZeroSFPoly, since they are both optional, so we would want it to work both ways to be completely generic.
If you make it consistent across predicted/measured, and allow for the scenario where SigmaZeroSFPoly is not populated, I will approve merge request.
|
My mistake @utwade, I misunderstood the content of Radiometric.NoiseLevel.NoisePoly. I'll leave this as a TODO, but not sure I have the bandwidth right now to add full support for calibrated noise measurements. Thanks for your response. |
No description provided.