stitching problems... (weird glare)

[Carles]

Hi all,

so I had been doing 3 panel mosaic of the Veil nebula, around 6hours each panel with L-extreme filter.

Thing is, individually the images can be wiped,stretched and so on without problems
but if I try to stitch it (with ICE as I am not aware of a better free option..) there's some
weird glares or color distortions all around the image. specially between the central and the
two external panels.

Each panel is properly cropped (used AutoDev to highlight stacking artifacts and zoomed a lot to crop as precisely as possible.. )
Also, stacking, in Siril, but previously registered and selected best frames in DSS based on Score and Star count.
This is default Wipe

Screenshot-2.jpg (319.88 KiB) Viewed 4880 times

More aggresive settings

Screenshot-3.jpg (422.76 KiB) Viewed 4880 times

And basic stretch in FilmDev to bring out the problem..

Screenshot-4.jpg (304.56 KiB) Viewed 4880 times

[Carles]

And following are the individual Panels in Wipe with default settings.
they seem OK to me, so... is ICE introducing artifacts when stitching?
I do the stitching in Linear state (so just open in StarTools, AutoDev, Crop, Restore->original)

Screenshot-7.jpg (404.44 KiB) Viewed 4879 times

Screenshot-6.jpg (383.15 KiB) Viewed 4879 times

Screenshot-5.jpg (414.65 KiB) Viewed 4879 times

Any Ideas?

Clears skies

Carles.

[Stefan B]

Hi Carles,

as far as I know you can stitch linear data in ASTAP. I know you had problems with this piece of software for stacking but you can stitch stacks in ASTAP which were stacked in another software I think.

Here's a suggested work flow with an example of DSS data at the bottom:

https://www.hnsky.org/astap.htm#mosaic

I haven't tried it but would like to if the weather permits any time soon.

Another possibility might be to process each stack seperately, get rid of the individual gradients and stitch afterwards in ICE to see what ICE comes up with.

If you want to try an option which is semi free you might want to have a look at a trial version of APP. I've heard a lot of good things about its mosaic composite capabilities. At least you could try for free. If you like it you will have to pay in the future of course...

Good luck!
Stefan

[Carles]

Hi Carles,

as far as I know you can stitch linear data in ASTAP. I know you had problems with this piece of software for stacking but you can stitch stacks in ASTAP which were stacked in another software I think.

Here's a suggested work flow with an example of DSS data at the bottom:

https://www.hnsky.org/astap.htm#mosaic

I haven't tried it but would like to if the weather permits any time soon.

Another possibility might be to process each stack seperately, get rid of the individual gradients and stitch afterwards in ICE to see what ICE comes up with.

If you want to try an option which is semi free you might want to have a look at a trial version of APP. I've heard a lot of good things about its mosaic composite capabilities. At least you could try for free. If you like it you will have to pay in the future of course...

Good luck!
Stefan

Hi Stefan!

thank you for your reply I will try the ASTAP method. I've tried processing them separately, then stitch in ICE but still the outcome was weird... I've downloaded the APP free trial this morning, but I assume it needs a powerful PC...it took me really long time and I gave up for now (plus not really know what I was doing..)

thank you for your suggestions! much appreciated it!

CS
Carles.

[Cheman]

Untitled (Medium) (Small).jpg (326.96 KiB) Viewed 4862 times

could the problem be related to the reflection from a star out of the field of view. It looks like it could be coming from Fawaris III. Ive had this problem before

[Stefan B]

Carles, would you mind sharing your three stacks? I probably can't fix your problem but I'd like to practice a bit with real world data as long as I can't image myself

CS
Stefan

[Mike in Rancho]

Carles, would you mind sharing your three stacks? I probably can't fix your problem but I'd like to practice a bit with real world data as long as I can't image myself

CS
Stefan

I'd love to give it a go also!

I've only really done one mosaic at all - the Dragons from the big Elf practice thread on CN. It was only two panel, but each panel was separate HaRGB!

Cropping and registering were trouble enough, but I could not replicate the (PI) workflow that was being suggested using my ST and DSS. The different gradients and brightness from all the different filter sessions just would not match well, and of course each required differing levels of it. And ST of course personalizes Wipe and AutoDev to every image.

In the end a linear stitch just wouldn't work out. So I processed each panel in full separately, using saved references from one panel to try to match up the level of Wipe, AutoDev, and color balance on the second panel. Then I took those final outputs into Gimp to line them up and created a gradual layer blend of the overlap area. Imperfect and a fair bit of labor involved, but it seemed to work. 3 panel though, yikes!

[ionia23]

Even with some of the issues being reported, this Image Composite Editor program might be what I've been looking for. Yonder camera has a very small field of view. I can barely fit 1/4 of, say, M31 into one frame. Rather than using some ridiculously overpowered focal reducer, I could make a really impressive mosaic that would be quite high in resolution without being oversampled.

If I ever get a cloudless night again!

[Cheman]

ICE does a very good job, is easy to use and being free, cant be beat,in my experience
Che

[Carles]

Thanks all for your messages

at the end, I analised the subs again with ASTAP (which kinda gives me more clues than DSS..just my impression) and stacked with Siril. Just in case, did background extraction in Siril, then stitched in ASTAP and processed in StarTools as usual

Screenshot-2.jpg (332.68 KiB) Viewed 4775 times

not perfect, but the weird stuff went away...

Cheers

Carles.