Jump to content

Stargazers Lounge Uses Cookies

Like most websites, SGL uses cookies in order to deliver a secure, personalised service, to provide social media functions and to analyse our traffic. Continued use of SGL indicates your acceptance of our cookie policy.

Rodd

Rosette Experiment

Recommended Posts

I have imaged the Rosette Nebula 3 times at 3 differnt focal lengths (318 mm, 480 mm and 700 mm).  I managed to work each idata set into a version I could live with (after much time!).  Each image consists of 14 hours, 24 hours, 13 hours respectively.  I wondered what would happen if I combied them all into a 51 hour image.  So, becuase I do not know how to insert higher resolution data into lowere resolution widefield data (Any advise on how to do that would be very much appreciated).  So, I registered all data to the smallest FOV (which eliminated the need to crop the combined image), combined the 3 images,then inserted the lum of the highest resolution image (700 mm)-- Well--here it is.  I do miss the wider FOV, but still a satisfying endeavor.

 

XYZc.thumb.jpg.3e3ac2b4c403c63f8421eeb103278a9b.jpg

Edited by Rodd
  • Like 42

Share this post


Link to post
Share on other sites

Well, you've certainly picked up mases of detail there, one could spend ages just exporting the Bok globules alone.

Share this post


Link to post
Share on other sites

That is absolutely stunning.

Share this post


Link to post
Share on other sites
56 minutes ago, DaveS said:

Well, you've certainly picked up mases of detail there, one could spend ages just exporting the Bok globules alone.

Thanks Dave!

Rodd

Share this post


Link to post
Share on other sites
46 minutes ago, Skyline said:

The colour rendering is wonderful.

Thanks Skyliner. 
Rodd

Share this post


Link to post
Share on other sites
31 minutes ago, MarkAR said:

That is absolutely stunning.

Thanks Mark

Rodd

Share this post


Link to post
Share on other sites
2 hours ago, simmo39 said:

AMAZING!

Thanks Simmo!

Rodd

Share this post


Link to post
Share on other sites

A great image, well done.

Bob.

Share this post


Link to post
Share on other sites
6 minutes ago, bob-c said:

A great image, well done.

Bob.

Thanks Bob!

Rodd

Share this post


Link to post
Share on other sites

Well Rodd, we can conclude that experiment worked a treat - it is a superb final image, the detail is brilliant.

Share this post


Link to post
Share on other sites
32 minutes ago, Miguel1983 said:

amazing, 3D-like 👌

Thanks Miguel

Rodd

Share this post


Link to post
Share on other sites
32 minutes ago, geoflewis said:

Well Rodd, we can conclude that experiment worked a treat - it is a superb final image, the detail is brilliant.

Tha Geof

Rodd

Share this post


Link to post
Share on other sites

I integrated the images a bit differently--before, I used Pixle Math to combine them.  In this case I used the integration tool and combined the 3 by the average method.  i think it has a bit more depth

XYZ-New4.thumb.jpg.89a5ca7fae1bc77ba5b2c52b39ef1480.jpg

  • Like 2

Share this post


Link to post
Share on other sites

Very subtle changes Rodd, but the slightly darker shadows do give the image more depth and I like it. That said, if ever I produced an image as good as your V1 I wouldn't touch it ever again.....

Share this post


Link to post
Share on other sites

Scrumptious. I'm a great fan of blending higher resolution 'areas of interest' into widefield images of lower resolution. I do this using Registar followed by Photoshop. Usually I'm working with processed data because I already have it. Starting at the linear stage all round is a luxury. This is what I do:

1) Open Hi Res and Wide in Ps and try to get the Hi Res looking as if it could be blended in. (Adjustments to levels, colour balance, etc., just by eye.

2) Take both into Registar and Register Hi Res to Wide. Then Calibrate Hi Res to Wide. Then Crop and Pad Hi Res to wide and save it. I don't combine them in Registar. The terms in red are standard functions in Registar, one click for each.

3) Import the Hi Res output from Registar into Ps and paste it as a top layer onto Wide. It will be perfectly aligned with a black 'padding' border round it which I discard by using Colour Select set to zero fuzziness to locate it and Delete to discard it.

4) Now the fun begins. Where there is a big difference in resolution (as there usually is in my case) I want to merge the Hi Res in slowly so that its outer edges will be invisible and its contribution to the bottom layer will increase towards its centre. I use a feathered eraser to let it in slowly. Adjustments to brightness can be done using Dodge and Burn brushes and individual channels can be given this treatment to get a seamless blend. Finally the percentage opacity of the top Hi Res layer is something I judge by eye based on looking seamless and having the lowest noise (though a big downsizing of the Hi Res usually eliminates noise completely in my case. As you can see, this is a hands on, 'artisitic,' process and not a uniquely digital one. I love doing it but some would hate it.

Some examples:

https://www.astrobin.com/cqd5z8/0/?nc=user

https://www.astrobin.com/335042/?image_list_page=2&nc=&nce=

https://www.astrobin.com/321869/?image_list_page=2&nc=&nce=

If the differences in resolution are not great you might just get away with combining the linear sets with hard edges but I wouldn't bank on it. I suspect it will always be an arty process!

Olly

PS It makes perfect sense to use all the data you have. Running an imaging guest house means I often have multiple datasets and in my view it makes sense to combine them as well as to to make an image from the new data. (Naturally guests want to do that because it's their primary image but why not also make a stack of stacks?)

 

Edited by ollypenrice
  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

I can't imagine anything looking better than this. Outstanding!

Share this post


Link to post
Share on other sites
3 minutes ago, ollypenrice said:

Take both into Registar and Register Hi Res to Wide. Then Calibrate Hi Res to Wide. Then Crop and Pad Hi Res to wide and save it.

Olly, I started using Registar about 18 months ago after several recommendations and a quick tuition session by @carastro, but I only use Register and Crop/Pad, so what does calibrate do?  I don't think I've even noticed that function previously.....

Also I can't find much in the way of tutorials for Registar, so if you know of any please could you share them.

Cheers,

Share this post


Link to post
Share on other sites
3 minutes ago, geoflewis said:

Olly, I started using Registar about 18 months ago after several recommendations and a quick tuition session by @carastro, but I only use Register and Crop/Pad, so what does calibrate do?  I don't think I've even noticed that function previously.....

Also I can't find much in the way of tutorials for Registar, so if you know of any please could you share them.

Cheers,

Always edge crop anything you take into Registar. Vital.

You open two images in Registar and register, let's say, B to fit A.   B will then be reorientated and resized to fit A but it will not be adjusted in any way to 'match' B in terms of levels, colour balance, contrast etc. Only if you activate Calibrate will it be adjusted in order to make a seamless blend with A.

If working with linear data you have a fighting chance that the two images will now match well enough for a seamless blend via Registar's Combine Images command. But even with linear data there may be a line visible. Don't panic! Read on...

Let's say images A and B are still linear and framed to make a mosaic, A on the left, B on the right. Ask Registar to register and calibrate B and then combine them. Save the combined image but also save the registered calibrated version of B on its own. You may need this as a 'patch' panel.

In Photoshop stretch the combined image not all the way but hard enough to reveal any defects in the calibration but record your stretching as an action. There are two potential sources of a visible line, the right hand edge of A or the left hand edge of B. Let's say we can see a line at the right hand edge of A (so visible on panel B of the combined.) This is sooo easy to fix...

Open the saved Registered Calibrated B and run your recorded stretching action on it. Now it will be all but identical to the stretched combined image. Paste it onto the stretched combined and move it into the right place with the move tool. Use a feathered eraser to take off all of this top patch panel except the bit covering the line or joint defect. Flatten and congratulate yourself!

If the visible line is from the left hand edge of B it is on the original A panel, which you already have. Run your recorded stretch and patch in the same way.

I don't know any Registar tutorials but in this post and the one above I think I've set down all the methods I've evolved in using it. PM me for clarification if necessary. I couldn't function without Registar.

Olly

 

  • Thanks 1

Share this post


Link to post
Share on other sites
38 minutes ago, ollypenrice said:

I don't know any Registar tutorials but in this post and the one above I think I've set down all the methods I've evolved in using it. PM me for clarification if necessary. I couldn't function without Registar.

Hi Olly,

I'll take my time to digest your above notes and see what I can do.

Many thanks,

  • Like 1

Share this post


Link to post
Share on other sites
4 hours ago, geoflewis said:

Very subtle changes Rodd, but the slightly darker shadows do give the image more depth and I like it. That said, if ever I produced an image as good as your V1 I wouldn't touch it ever again.....

Thanks Geof.  Yes...its subtle--I had to look back and forth for a time before I was convinced.  I was suprised there was any difference, really--I guess the algorhym is different.

Rodd

  • Like 1

Share this post


Link to post
Share on other sites
3 hours ago, ollypenrice said:

Scrumptious. I'm a great fan of blending higher resolution 'areas of interest' into widefield images of lower resolution. I do this using Registar followed by Photoshop. Usually I'm working with processed data because I already have it. Starting at the linear stage all round is a luxury. This is what I do:

Thanks Olly.  If I used PS I would deffinitely try it.  It seems in PI it is a bit trickier

Rodd

Share this post


Link to post
Share on other sites
3 hours ago, Mr Spock said:

I can't imagine anything looking better than this. Outstanding!

Thaks Michael!

Rodd

Share this post


Link to post
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

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.