Jump to content

Banner.jpg.b83b14cd4142fe10848741bb2a14c66b.jpg

Orion Constellation - 6 Pane Mosaic

Hytham

By Hytham
in Imaging - Deep Sky

Recommended Posts

Hytham

Hytham

Posted
  • Author
Posted

Thank you, everyone.

Gina, I don't know if a colour version is going to be in the works because it will take a great deal of time to complete, unless...

I have been thinking of purchasing a one shot colour CCD - or a mono ccd - and using that to collect my colour data simultaneously using a Canon 200mm lens (or something similar). This way I can collect higher resolution narrowband and luminance data with the main rig and the lower res colour data with the significantly wider lower res set up.

Always more money to spend, isn't there? Unless there is some one kind enough to collect that colour data for me ;)

Link to comment
Share on other sites
  • Replies 54
  • Created
  • Last Reply
ollypenrice

ollypenrice

Posted
Posted

It's difficult to say just how much of  a reduction in resolution for colour will work. Recently we tried using TEC luminance (FL980mm) over Tak 106 colour (FL530mm.) Initially it didn't look too good. The stars were the worst, looking bloated and coarse. I applied the luminance anyway and then did some more star reduction. It looked OK in a small format but a mess when seen up close. However, I then re-applied the luminance and at this point it started to look very promising.

We then shot a set of RGB from the TEC as well and there really wasn't all that much difference between the TEC-Tak and the TEC-TEC, so we combined them to make this;

http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-L8kQcbB/0/X3/IRIS%20GHOST%20TEC140%20WEB%202-X3.jpg

For me it depends on the image. Where I have lots of stars on dark background sky I like to make it from RGB only, without luminance. This needs full resolution colour, of course.

I don't have Orion in a 200mm lens but if you'd like to have a play with RGB captured at 85mm I can supply it, I think. It would be the RGB from this HaRGB widefield. PM me if you'd like it.

http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-V9SQcqS/0/X3/ORION%2085MM%20LENS%20HaOSC%206%20PANEL-X3.jpg

Olly

Link to comment
Share on other sites
Ewan

Ewan

Posted
Posted

Hytham it's astrophotographers like you that give us the gentle kick in the back side to do better, stunning, amazing, great etc does not give this picture justice at all & to top it off it's just in Ha 5nm :shocked:.

If the almighty wanted a photographer to document his work he'd be giving you a call dude :grin:

I had no idea this area was so rich in data, it is truly huge, probably one of the largest single targets in our night sky.

I was debating buying a narrower Ha filter & ccd to dslr adapter for shots like this so thanks to you my xmas prezzie has been decided.

I hope you manage to collect the rest of the data needed pretty soon as i & others here are dying to see the end result, i'm sure we will be blown away.

Love it.

Link to comment
Share on other sites
symesie04

symesie04

Posted
Posted

That sir is just simply magnificent. Really unsure what i can say to express the quality of that image, Detail amazing and wonderful field of view. It certainly would look incredible in colour but can see how time consuming it would be to remap the colour data. A joint project with some of the other great imagers on here sounds like a great plan though. Besides im hoping your just going to continue working your way out and producing mosaics of much more sky ;)

Link to comment
Share on other sites
Epicycle

Epicycle

Posted
Posted

Thank you, everyone!

Sarah, I use PI to stictch the panels together, but I do have a lot of problems using it.  The original registration is always dead set without producing any nasty artifacts along the sames, but the problem arises after I start to generate the registered frames for the Gradient Merge Mosaic tool.  Star Alignment's register/match module always ends up distoring the frames significantly along the seams causing a doubling effect and forces me to fix everything in post.  It's unbelievably annoying.  Even when I explicity use 2-D surface splines combined with distortion correction enabled, with the residual and iterations left at default and one final increase to the RANSAC from 2 to 8, I still have problems. 

I have not run into this issue when using the scope at a native F5 with the smaller FOV and a larger overlap.  I think the key for me is to increase my overlap from 10% to 20% giving the RANSAC algorithm a higher sampling producing a more accurate result.

In the event of a smaller overlap I select an area on each intersecting panel where the overlap will occur allowing the algorithm to perform its random sample consensus more accurately.  It's certainly not an exact science and still trying to learn how to do this stuff.

I tried PS and it just wasn't happy which was very surprising to me as the mosaic I was using to test against, PS had so little issue creating a beautifuly seamless mosaic with zero artifacting. 

Hi Hytham,

terrific image obviously. I had the same problem you describe creating a mosaic of 20 panes in PI. I ended up with about 3 panes which were

very stubborn. Surprisingly what helped was to rotate them and increase a little the distortion residual, but as you said it's not a very "exact science".

The idea of the generated star field mentioned in Sara's post sounds promising...It's seems to me all big mosaics will eventually face some distortion

problems so it's worth having an efficient method to deal with the issue,

cheers

E.

Link to comment
Share on other sites
Sandancer10

Sandancer10

Posted
Posted

Can't praise you enough - these are stunning images.  I have yet to look at the HH Nebula and I had no idea how large an area this Nebula covers, normally it's only the HH detail that is shown. I can't imagine what your finished project will look like - superb work.

:smiley:

Link to comment
Share on other sites
Hytham

Hytham

Posted
  • Author
Posted

Hi Hytham,

terrific image obviously. I had the same problem you describe creating a mosaic of 20 panes in PI. I ended up with about 3 panes which were

very stubborn. Surprisingly what helped was to rotate them and increase a little the distortion residual, but as you said it's not a very "exact science".

The idea of the generated star field mentioned in Sara's post sounds promising...It's seems to me all big mosaics will eventually face some distortion

problems so it's worth having an efficient method to deal with the issue,

cheers

E.

Thank you!  I have also found that the accuracy of the registration of panels is also partially dependent upon my ability to accurately generate a proper DBE model against each panel.  While testing against my first mosaic I had a number of problems, but after regenerating a proper model the distortion was acounted for properly. 

I have thought about using Registar to register each image and then carrying that into photoshop to bring all the pieces together.  Of course this brings up other problems that I potentially foresee and that is the varying degrees of brightness between each panel that will require manually equalization of brightness.  That doesn't sound fun, but something I'm willing to give a shot.

It's difficult to say just how much of  a reduction in resolution for colour will work. Recently we tried using TEC luminance (FL980mm) over Tak 106 colour (FL530mm.) Initially it didn't look too good. The stars were the worst, looking bloated and coarse. I applied the luminance anyway and then did some more star reduction. It looked OK in a small format but a mess when seen up close. However, I then re-applied the luminance and at this point it started to look very promising.

We then shot a set of RGB from the TEC as well and there really wasn't all that much difference between the TEC-Tak and the TEC-TEC, so we combined them to make this;

http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-L8kQcbB/0/X3/IRIS%20GHOST%20TEC140%20WEB%202-X3.jpg

For me it depends on the image. Where I have lots of stars on dark background sky I like to make it from RGB only, without luminance. This needs full resolution colour, of course.

I don't have Orion in a 200mm lens but if you'd like to have a play with RGB captured at 85mm I can supply it, I think. It would be the RGB from this HaRGB widefield. PM me if you'd like it.

http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-V9SQcqS/0/X3/ORION%2085MM%20LENS%20HaOSC%206%20PANEL-X3.jpg

Olly

Now that I have put more thought into it and actually did the math, the use of an 85mm lens with the KAF-8300, or a 50mm with the ICX 694, may provide me with the best resolution to easily stitch the images together to acquire the colour data.

I guess the question that really remains is it wotth going to the one shot colour, or should I continue down the route of mono? I already have the ATIK EFW2 FW with Astronomik RGB filters.

I'm starting to wonder if it's even worth collecting colour data using mono as eventhough it provides some detail, but that would end up requiring potentially longer integration times to compensate for the decrease in sensitivity. 

Everyone else, thank you very much for your kind words.  They are very much appreciated!

Link to comment
Share on other sites
ollypenrice

ollypenrice

Posted
Posted

Thank you!  I have also found that the accuracy of the registration of panels is also partially dependent upon my ability to accurately generate a proper DBE model against each panel.  While testing against my first mosaic I had a number of problems, but after regenerating a proper model the distortion was acounted for properly. 

I have thought about using Registar to register each image and then carrying that into photoshop to bring all the pieces together.  Of course this brings up other problems that I potentially foresee and that is the varying degrees of brightness between each panel that will require manually equalization of brightness.  That doesn't sound fun, but something I'm willing to give a shot.

Now that I have put more thought into it and actually did the math, the use of an 85mm lens with the KAF-8300, or a 50mm with the ICX 694, may provide me with the best resolution to easily stitch the images together to acquire the colour data.

I guess the question that really remains is it wotth going to the one shot colour, or should I continue down the route of mono? I already have the ATIK EFW2 FW with Astronomik RGB filters.

I'm starting to wonder if it's even worth collecting colour data using mono as eventhough it provides some detail, but that would end up requiring potentially longer integration times to compensate for the decrease in sensitivity. 

Everyone else, thank you very much for your kind words.  They are very much appreciated!

Do I understand your point? Why would mono RGB be less sensitive than OSC? I think they'd be roughly equivalent with mono having a slight advantage. The Bayer Matrix (I think) slightly diminishes pixel area and you are condemned to the daytime-orientated matrix of RGGB. Why use two greens for astronomy? I have seen an OSC SBIG 8300 in action and it was very sluggish. In fact the owner sold it after a short spell with it.

Olly

Link to comment
Share on other sites
Hytham

Hytham

Posted
  • Author
Posted

Do I understand your point? Why would mono RGB be less sensitive than OSC? I think they'd be roughly equivalent with mono having a slight advantage. The Bayer Matrix (I think) slightly diminishes pixel area and you are condemned to the daytime-orientated matrix of RGGB. Why use two greens for astronomy? I have seen an OSC SBIG 8300 in action and it was very sluggish. In fact the owner sold it after a short spell with it.

Olly

You're not because my explanation is not exactly clearly written.  My apologies for that.

The internal debate I'm having is that since RGB is not generally used to collect the very fine detail then going the route of OSC would be an advantage in its court.  Unfortunately the RGGB matrix makes OSC is less sensitive than RGB forcing integration times to be longer for the OSC camera to make up for the same signal acquired by RGB mono.  I have seen some amazing images out there produced by OSC and there is detail to be had, but ...

For simply grabbing colour and not worrying about detail, I *think* OSC may be ok.... maybe.

Link to comment
Share on other sites
Special K

Special K

Posted
Posted

Simply stunning! Smoking gun from the Big Bang. I love the black and white personally. Absolute wonder of the heavens in this image.

I was looking at photos of the Helix earlier today and Barnard's Loop is really reminding me of this in your mosaic.

If only we could observe that amazing LIGHT around Alnitak with the naked eye.

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.