Using flats on full disk solar images

Merlin66 (Ken) · #1 08-04-2021, 03:18 PM

All the info readily available on the use of flats (to suppress NR and dust bunnies) is applied to surface images and not to applying them to full disk solar images.
Why is still so?
Full disk images also suffer from similar “defects”.....
One very successful method of obtaining solar flats is to use the clear plastic bag over the aperture.
I’ve been trying this method with SER videos (Firecapture) one video of the full solar disk and one with the plastic wrap in place. Using AS3! Latest version V3.1.4. A master flat can be made as a tiff and then applied during the stacking of the solar disk. Sounds pretty straight forward.....
(Seems to work well with surface images)
After three days of trials, I found a major problem.
Summary: For full disk flats to work, the master flat image must be registered 100% with the solar disk stack/ individual frames. There’s no way I can find to actually do this! It really comes down to having ABSOLUTELY PERFECT Polar alignment with NO drift between the flat image and the solar disk stack/ individual frames.
Obviously my alignment wasn’t 100%
This is not a major problem when normally imaging, AS3! will make the necessary minor corrections to give a quality outcome. The problem is how to register the flat to the individual frames ( the flat is applied to each frame during the stacking process). I don’t know the answer, but I do know if you re-align the mount with some accuracy there is enough registration to remove NR etc.
I have some sample images showing the results. Still not perfect, but certainly an improvement in the suppression of NR.
If anyone has found a better more robust solution I’d love to hear about it.

Onwards and upwards.

h0ughy (David) · #2 08-04-2021, 05:35 PM

Your brave I was never successful with any flat and gave up. PS clone was used a few times for very bad dust bunnies. I did however try to perfect cleaning and sealing the image train as much as possible and cleaned the objective before a image session. It cut down so much handling time after the fact

Merlin66 (Ken) · #3 08-04-2021, 05:53 PM

Houghy,
Dust doesn’t seem to be my primary issue, it’s the annoying NR.

Merlin66 (Ken) · #4 12-04-2021, 10:31 AM

Limited success.
Marty (SolarChat) has suggested the flats need to be more diffuse...
I'll re-try using tracing paper (one layer or more...)
Onwards and slightly upwards

h0ughy (David) · #5 12-04-2021, 10:56 AM

Oh so that also removes Newton rings?

Merlin66 (Ken) · #6 12-04-2021, 11:17 AM

David as they say...."That, is the right question!" (I Robot)

Unfortunately the Marty examples (so far) don't seem to have been applied to solar disk images showing NR.

Merlin66 (Ken) · #7 27-04-2021, 04:19 PM

I've finished the first draft of my re-write up.
The DropBox link:
https://www.dropbox.com/s/fxsjvi20qr...l2021.pdf?dl=0

Comments/ feedback welcome.
Summary:
A good choice of "plastic bag" as a flat diffusion filter works well with the FireCapture "Flat" - which is then applied to each video frame. Stacking in AS3! completes the process.