[News] sub window statistics

[Ewout M. Helmich]

Dear all,

We would like to finalize the procedure for deriving and storing 
statistics in
subwindows for the images, and it turns out we're still somewhat unsure 
whether the
current approach is in fact the best way to do things.

Currently the SubWinStat situation is that for each CCD statistics are 
calculated
for:

A) 32 subwindows covering the entire CCD. The windows are 500x500 pixels 
in size.

Our worry is that with this subwindow size, large scale (flat-field) 
effects and
stars may be a factor within the subwindows. Additionally when the idea 
is to
characterize the CCD images by a reduced number of numbers, 32
subwindows may be counter productive, complicating things rather than 
simplifying
them.

We could use smaller and fewer windows, for example:

B) 8 subwindows NOT covering the entire CCD. Windows are 50x50 pixels in 
size.

What are your ideas on this?

Regards,
Ewout