Jump to content

Banner.jpg.b6007b69ccdf5c69bf18273ddfe023df.jpg

Mosaic help please


Recommended Posts

Posted (edited)

If anyone thinks that one system will always work for deep sky, they haven't done enough mosaics, or they've been luckier than I have! One system will not always work.

If you're lucky, you can combine calibrated, flattened, gradient-corrected, colour-balanced linear images and get a seamless result. I find that this occasionally happens on modest mosaics, up to maybe six panels, though it's far more likely that there will be the odd joint or glitch visible when you stretch. There is a quick way to test the seamlessness of your mosaic: run it through the Equalize adjustment in Photoshop. Joints, colour changes, etc etc, will scream at you! Don't panic, this is a diagnostic tool to tell you where your problems lie. You'll then discard it.

The most consistently reliable method I've found for very large mosaics  (like the 30-odd panel ones I worked on with Tom O'Donogue and Yves Van den Broek) is to combine partially stretched panels using Registar. However, keep all the registered-calibrated panels that went into this. Move on to Photoshop and check for irregularities using Equalize. They will be there. Use the saved registered-calibrated panels, adjusted by hand, to make patches over the problem areas.

The software to make seamless linear deep sky mosaics reliably, such that they can then be stretched and processed for local contrast enhancement, etc, which you will want to do, does not exist.

Olly

Edited by ollypenrice
Clarification
  • Like 1
Link to comment
Share on other sites

32 minutes ago, ollypenrice said:

The software to make seamless linear deep sky mosaics reliably, such that they can then be stretched and processed for local contrast enhancement, etc, which you will want to do, does not exist.

Olly

True, but the PI Photometric Mosaic script has a darn good go at it.

Link to comment
Share on other sites

11 hours ago, tomato said:

True, but the PI Photometric Mosaic script has a darn good go at it.

That's good to know. Many people report success with APP as well but in this 32 panel, captured by Yves Van den Broek, it was able to make an excellent job of the geometry but not of the seamless joining or calibrating. Yves is very competent with software but short of time, so he gave it to me to assemble the DBE subtracted panels. I used the DBE- partial stretch-manual adjust- join-patch-check in PS Equalize-repeat method described above. It isn't fast! I worked in strips of 5 images at a time and then joined the strips together.

https://www.astrobin.com/g82xf7/B/

The trouble is that are are things the software won't automatically allow for, like changes in FWHM, levels of sky transparency affecting levels of blue bloat, etc etc. With a large mosaic you can't hope for perfect or identical sky conditions for every panel. Far from it. Some of the panels in the Cepheus image needed quite a lot of cosmetic work to make them blend in visually with the rest.

Almost everything I now do is a mosaic and sometimes, but only exceptionally, a linear combining of DBE-subtracted, colour-calibrated panels gives a stretchable result. Of course, this also depends on how hard you want to stretch but it's in the nature of mosaics to beg for very hard stretches in search of the faint, linking nebulosity between 'areas of interest.'

Olly

  • Like 1
Link to comment
Share on other sites

@Laurin Dave suggested trying the Photometric Mosaic script to stitch together a 12 panel mosaic of M31, after I had failed miserably to produce a decent result with APP. The panels were captured over many weeks and had widely varying background qualities, but by using the script’s method of employing a widefield reference image, he was able to produce a result far in excess of my expectations, based on the data.

CF964A78-DB50-4D25-8091-5F334B78D046.thumb.jpeg.9197fe617808904af5f7f8337199bd7d.jpeg

8ACB69E6-14FE-486B-BE8D-208139250A47.thumb.jpeg.f4a58433d23fd79d248cc728b207feb9.jpeg

  • Like 1
Link to comment
Share on other sites

On 27/05/2022 at 09:07, tomato said:

@Laurin Dave suggested trying the Photometric Mosaic script to stitch together a 12 panel mosaic of M31, after I had failed miserably to produce a decent result with APP. The panels were captured over many weeks and had widely varying background qualities, but by using the script’s method of employing a widefield reference image, he was able to produce a result far in excess of my expectations, based on the data.

CF964A78-DB50-4D25-8091-5F334B78D046.thumb.jpeg.9197fe617808904af5f7f8337199bd7d.jpeg

8ACB69E6-14FE-486B-BE8D-208139250A47.thumb.jpeg.f4a58433d23fd79d248cc728b207feb9.jpeg

I have been impressed by the PI Photometric Mosaic method. It has worked well for me, although I have never tried anything as big as this! So it is now my first choice method.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.