[daip] A suggestion to improve putting boxes in MR CLEAN in AIPS task IMAGR

[Subhashis Roy]

Hi,

I have one suggestion to improve IMAGR multi facet multi resolution
CLEAN:

Often the extended diffuse sources are not fully inluded in a single
facet while using IMAGR.  Consequently, their sidelobes also extend
along many facets.  As you know, the invention of BOX in Imaging was to
confine the solutions. However, during the actual process in MR clean,
it becomes really difficult to realise whether the emission in one such
facet shown on the TV is the actual source or its sidelobe, and whether
CLEANing is reducing the sidelobes of extended sources. 
(This is important as interferometers often miss the required short
spacings, and I have noticed it while imaging the Galactic Centre with
the GMRT data)

A possible solution will be the following:

After we call 'Remake Images' and residual images are made, make one
more set of images, which are undersampled copies of the low resolution
facets. Then call FLATN (user can have the option), join them together
and show it on the TV.  User can have option of defining boxes on the
undersampled images of facets, which needs to be converted to
co-ordinates and written to a new BOX file, to be used subsequently by
IMAGR. Afterwards, CLEAN continues in the standard fashion.
I hope this won't require touching the code of IMAGR much.

Thanks a lot.

Subhashis