Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Creeping... Slowly, Inexorably, Relentlessly... forward

Uranium235

By Uranium235
in Imaging - Deep Sky

Recommended Posts

Uranium235

Uranium235

Posted
Posted

Well. its cloudy and im bored - so I've just uploaded the current state of play regarding my Cygnus mosaic.

Also, I've decided to increase the overall coverage to 20 panels, it doesnt really need to go past 16 - but I want it to sit nicely on a 16:9 monitor once Ive filled the gaps Im going to do another four down the left hand side - which should take it up to about 120 megapixels. Also, it would line up the rest of Cygnus quite nicely becuase if I were to do another 20 below this, it would bring in the NAN.... but not this year, there just isnt enough time left on it.

But after the first 20 are done, I'm going to experiement with adding colour via binned OIII data.... yes, I know thats bonkers with a Star71 but I could eliminate the "blocky stars" issue simply by removing them! The only exceptions would be the Tulip and Crescent, both of which I could improve by using the 200 or 130pds to shoot at higher resolution. Come to think of it, I already have 130pds OIII data for the crescent just laying about doing nothing.

Finally, I will go over the whole area with a DSLR at a low(ish) ISO to get some star colour back.... well thats the plan anyway! :)

 

29736595091_ca05bfdb0d_k.jpg

Link to comment
Share on other sites
  • Replies 59
  • Created
  • Last Reply
Uranium235

Uranium235

Posted
  • Author
Posted
28 minutes ago, uhb1966 said:

Gee... great! It is prob'ly a noob question but - what software do you use to get the panels so perfectly to match? There's no difference in exposure , sharpness... nothing!

Registar does the initial hard work, then over to Maxim to stitch them together.

Originally, I had a problem with differing background levels but a thanks to a handy hint from Olly, I fixed that by using normalisation when stacking. Also, you have to be slightly ruthless in what subs you keep - as anything that might damage the project needs to be chucked out.

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
1 minute ago, Ibbo! said:

Very nice.

I am doing a samll bit of Cygnus and learning that planning a mosaic is a big part of actually getting one finished.

Great stuff Rob I hope you get all the data you want.

Thanks Steve :)

Yep, planning is an important part of it (where to start/end, can time be saved by rotating the camera etc..). I worked it all out in stellarium beforehand, but unfortunately I dont have the luxury of SG pro or plate solving so for me its all Artemis markers and user locations in the handset.

Link to comment
Share on other sites
PhotoGav

PhotoGav

Posted
Posted

Give the man a medal! This is superb. You must be completely deranged to be able to keep your patience with a project of this colossal size. I am working on a four pane mosaic and it's pushing me to the very edge of my capabilities, both technical and mental.

Please share your technique for getting the panes to match so seamlessly... You mention an Olly trick of normalisation - please explain. I run a normalise routine in Nebulosity4 on all the subs of one pane before stacking them, but, I'm guessing here, is the trick to run the normalise across the (in my case 4) different stacked panes before taking them into Registar for stitching?

As for forcing Uranium Jnr to do a mosaic at the age of 6, I will have to report this to the NSPCC...

Link to comment
Share on other sites
Martin-Devon

Martin-Devon

Posted
Posted

Yes, I'm 'all ears' as well to hear about this normalisation step during stacking to even out the backgrounds of panels for a mosaic. To date I just eyeball the panels next to each other and adjust one or the other very slightly with curves in PS to get a match, but sometimes it plays up.

Martin

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
3 hours ago, PhotoGav said:

Give the man a medal! This is superb. You must be completely deranged to be able to keep your patience with a project of this colossal size. I am working on a four pane mosaic and it's pushing me to the very edge of my capabilities, both technical and mental.

Please share your technique for getting the panes to match so seamlessly... You mention an Olly trick of normalisation - please explain. I run a normalise routine in Nebulosity4 on all the subs of one pane before stacking them, but, I'm guessing here, is the trick to run the normalise across the (in my case 4) different stacked panes before taking them into Registar for stitching?

As for forcing Uranium Jnr to do a mosaic at the age of 6, I will have to report this to the NSPCC...

I'll put a screenshot of Maxim up later on, its combine tab (when using Sigma clipping, SD mask or Median) has a normalisation option for Linear or Delta Level. Set it to delta, and it pushes the histogram hard left after stacking - but not so hard as to clip the black point. I understand the same options are available in AstroArt, but not in DSS sadly.

Link to comment
Share on other sites
martin_h

martin_h

Posted
Posted
20 hours ago, Uranium235 said:

Thanks Steve :)

Yep, planning is an important part of it (where to start/end, can time be saved by rotating the camera etc..). I worked it all out in stellarium beforehand, but unfortunately I dont have the luxury of SG pro or plate solving so for me its all Artemis markers and user locations in the handset.

I lined all mine up by eye, but made a grid wall chart to keep track of things, letters across the top A,B,C etc and numbered down 1,2,3 etc. So the Crescent was G7 (my starting point). I also kept notes of the RA/DEC of each panel incase I had to go back to it, and as a starting point for another nights imaging.

 

 

IMG_20160921_185533.jpg

12873673_10156703394970475_575489043_o.jpg

Link to comment
Share on other sites
ollypenrice

ollypenrice

Posted
Posted
21 hours ago, uhb1966 said:

Gee... great! It is prob'ly a noob question but - what software do you use to get the panels so perfectly to match? There's no difference in exposure , sharpness... nothing!

He's good, isn't he??

This is a glory. The large scale structures are relentless. Keep going.

Olly

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
38 minutes ago, martin_h said:

I lined all mine up by eye, but made a grid wall chart to keep track of things, letters across the top A,B,C etc and numbered down 1,2,3 etc. So the Crescent was G7 (my starting point). I also kept notes of the RA/DEC of each panel incase I had to go back to it, and as a starting point for another nights imaging.

I remember that image well mate, its what I had in mind when starting this project. But, it also demostrates exactly what Tom OD and I were talking about when we met at the IAPY awards - in that the sky sort of behaves like the inside of a dome (or planetarium), and tapers inwards towards the zenith. Such are the issues when trying to put a curved sky onto a flat chip.

If I remember rightly, you did that with an 80ED? So hats off to your sir! :)

Link to comment
Share on other sites
martin_h

martin_h

Posted
Posted

ED80/QHY9, all good fun! And yes the curve was a pain but interesting at the same time. That image at full res worked out to about 22 x 16 feet....... Took ages to scroll across the screen in Photoshop. In fact I used to load it then reduce it to 50% to work on it.

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
17 minutes ago, martin_h said:

ED80/QHY9, all good fun! And yes the curve was a pain but interesting at the same time. That image at full res worked out to about 22 x 16 feet....... Took ages to scroll across the screen in Photoshop. In fact I used to load it then reduce it to 50% to work on it.

Thats exactly the issue Ive ran into (processing power... or lack of it!). Im still processing at maximum resolution before downsizing for the internet, otherwise Ps has a bit of a fit when I try to use the jpeg export function. It will do it.... but just very slowly!

 

For those who asked about the stacking and stitching here is what I was on about:

 

1) Normalisation in stacking (Maxim)

Maxim-stacking.jpg

 

2) Stitching in Maxim... I try to avoid the "Snap" function on large images as there is no need to do that once Registar has worked its magic, all that needs doing is a few nudges with the arrow keys to line up the stars. Most of the time you can leave the background equalization at "Auto", but every now and then I use "Manual" to enter my own value (plus or minus) to get the best blend possible. MS ICE is alright on images that are 2 - 4 panels, but for something this big I always go for Maxim.

 

Maxim-mosaic.jpg

 

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
1 hour ago, ollypenrice said:

This is a glory. The large scale structures are relentless. Keep going.

 

Youre the one to blame for this matey! :D

I was happy with propellor to crescent, but you planted the idea of going over to the Tulip.... seems like I made the right choice in the end though.

Link to comment
Share on other sites
Uranium235

Uranium235

Posted
  • Author
Posted
28 minutes ago, martin_h said:

I used MS Ice to stitch all of mine. Drag all 60 odd panels in to it and go and make a coffee.

I found ICE a little hit and miss when it came to doing large mosaics, sometimes it stitched stuff in completely the wrong order (even though the filenames were sequential) and would end up with some really wierd results. The only way I got around that was to combine to strips, then combine the strips (or do it in quarters) - a bit of a faff :)

Or maybe im just a bit of an OCD control freak, and want complete manual control over the stitching process...lol :D

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • 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.