Issues with computation of scans for synoptic maps

[detoma]

A preliminary analysis of UCOMP intensity scans for synoptic maps indicate that:

a) using a larger bins of 1.0 degree instead 0.5 degree gives better statistic

b) using the mean instead of the median to compute the average is very sensitive to outlier values and results in unwanted spikes

c) negative and zero values must be masked out before computing the average

d) a maximum intensity value is recommended to filter out hot pixels or unusually high intensity that are not physical

I suspect some/all of these findings apply to KCor, especially in the outer fov.