Using symmetries with DBE for lkarge nebula objects

[lucchett]

Hello,
I wonder if some DBE expert can help me with this problem.

I have an image of M16 that has been acquired at 20-25 deg elevation and shows obvious gradient.
I have a FOV of 2x2 degrees and the area is almost covered by the nebula.

the gradient is different in intensity for the diffefrent filtrers but it seems geometrically quite simple:

parallel to the lower edge, that is brighter, and going darker versus the top edge.
the gradient is probably shifted a bit on the upper half of the image.

so, my questions:
-shall I keep the main simmetry point offered by DBE?
-if yes, shall I move it a bit on the upper part, where the gradient start to be more noticeable?
-simmetries: do I need to run VERTICAL or run  HORIZONTAL symmetries?

this is the key point: running a VERTICAL means all the points on the line should have the same Background value, right?

-simmetries: only the start and last point of the symmetry line counts?
-additional points: is it ok to add also isolated samples where possible?

I can post the image on the challenge section, but I would like first to understand better the tool and see if I can manage.


Thank you very much in advance,
Andrea

[Carlos Milovic]

You should use Horizontal symmetries. Only the end point is considered. The starting point is considered as a valid sample, so it should be free from nebulosities. Move it if you feel that it helps you to get a better coverage of the image.
More samples are fine. Just don't over do it, and don't place samples over objects. In the worst case scenario, the samples should be on dark nebulas, and aim to have them all at the same level at the end.

[lucchett]

thank you Carlos, very helpful.
to be sure,
just one more question on the "starting point": is this valid also for the default symmetry point put by DBE in the middle of the field?

Are the central cross point and/or the two lines contributing to the model?

I mean, besides simmetry, it does sample the background?

I almost always added points to the initial default dbe "cross", but now I suspect this is a big mistake...


Thank you,
Andrea

[Carlos Milovic]

No, it is not a symmetry point. It is just the center of both axis. You may move it if you want. No sample is being taken there.

[lucchett]

good.
thanks a lot for the clarification
andrea