Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

The Iris in RGB


peter shah

Recommended Posts

NGC7023 Imaged over two nights and plagued by satellite trails. Imaged before moon rise on Monday 29th and Tuesday 30th nights. Exposure times were 8x 900s in Red 8x 630s in Green 8x 820s in Blue. Tricky one to process, pushing it so close to the noise levels really brings out faint gradients. Processed in Photoshop.

 

Click for full res but dont look too close ?

 

lEjLriQ_ilxa_16536x16536_wmhqkGbg.jpg

Link to comment
Share on other sites

  • Replies 38
  • Created
  • Last Reply

Like it very much.

I do have one thing that in my view sort of "out of place", and I've seen this a lot in images such as this one. It might as well be matter of taste (and as such - not up for debate) - but I think that dark nebulae in images tend to be "overexposed" in post processing. This way it looks more like emission type gas that glows rather than being something that obscures other things and is very dark in nature (not much light bounced of it at all, and its not glowing by itself).

I also think that this particular image would benefit from that kind of processing - it would bring in more "drama" into the image - "place where things shine almost hidden in the depths of universe" sort of feel ...

Link to comment
Share on other sites

Thank you everyone for your comments :thumbsup:

16 minutes ago, vlaiv said:

Like it very much.

I do have one thing that in my view sort of "out of place", and I've seen this a lot in images such as this one. It might as well be matter of taste (and as such - not up for debate) - but I think that dark nebulae in images tend to be "overexposed" in post processing. This way it looks more like emission type gas that glows rather than being something that obscures other things and is very dark in nature (not much light bounced of it at all, and its not glowing by itself).

I also think that this particular image would benefit from that kind of processing - it would bring in more "drama" into the image - "place where things shine almost hidden in the depths of universe" sort of feel ...

I think most of this lark is down to taste and interpretation and in my opinion it is art more that a science. That said I do try to keep the process even and keep it as a representation of what is actually there. There is a definite  reddish brown colour to the dark stuff.  I think its as you pull the dark stuff out of the background using curve stretches and push the saturation it naturally lifts the colour. Its how far you want to or can push the data. I wont purposely change what is there. 

6 minutes ago, cfpendock said:

That is exceptionally nice.  What scope / camera did you use?

Chris

Sorry Chris forgot this information.....My camera is a H35 by Starlight express and my Scope is an AG12

Link to comment
Share on other sites

Thank you Chaps and Chapesses :thumbsup:

42 minutes ago, swag72 said:

I don't know how you do it with such small amounts of data.....

Thank you Sara....Quite often I will pre-process and see how far I can go with it. More subs would make it so much easier.....This one was a challenge to process, as I got to the fainter stuff the gradients really started to cause me problems.

Link to comment
Share on other sites

Just now, tooth_dr said:

Absolutely stunning Peter, inspired by this I'm going to aim for this target first tonight.

Can I ask was there any specific reason for the seemingly random exposures lengths for the green and blue?

Not Random they are G2V star calibrated exposure weights to help with making colour balance easier.... I am able to apply the same curve stretches to each channel and the colour will very close to where it should be.....

Link to comment
Share on other sites

1 hour ago, peter shah said:

Not Random they are G2V star calibrated exposure weights to help with making colour balance easier.... I am able to apply the same curve stretches to each channel and the colour will very close to where it should be.....

Interesting, what did you use to calculate this?

Link to comment
Share on other sites

Stunning image but, rather than just chuck out a subjective value judgement like that, I'd like to be specific.

Vlaiv's point is a serious one. However, we have the technology to make the dust a 'region of interest' and to find within it variations in illumination and subtleties of structure which, surely, are worth exploring. OK, so we do this at the expense of its 'true' relationship with the more-or-less vacuum of the background sky but why not? How interesting is that background? I prefer Peter's dust. The point of this image is to lift it into visibility. Also, it takes no inversions of the histogram to render the dust brighter than the background. It is brighter. Simple as that. But the point of stretching is, of course, to exaggerate that extra brightness.

If anyone but Peter had posted 'Iris in RGB' I'd have been twitching to say, 'No no no, for the dusty stuff you need loads of luminance.' How Peter manages without it in this way, I have no idea. I'm agreeing with Sara here, I think. Well, hats off, Peter. I need yonks of luminance and you don't!

And those elusive pinks are coming through - in an unreasonably short total exposure time. Humph!!!

Respect.

?lly

Link to comment
Share on other sites

30 minutes ago, HunterHarling said:

Interesting, what did you use to calculate this?

Its quite a simple process. you just find a G2V class star near by and do three equal exposures through each RGB filter making sure the star is not saturated. Then measure the ADU of the G2V star and them divide one into the other to get your exposure weights. The ratios can be applied to any amount of exposure length. 

 

Link to comment
Share on other sites

As others have said: great image.

4 hours ago, peter shah said:

Click for full res but dont look too close ?

Of course I had to do just that ?

There seems to be a satellite trail in the blue master. If you have PixInsight, you could try large scale pixel rejection. This way you don't need to increase the rejection parameter in pixel rejection (and lose valuable data), and still get rid of this trail.

Link to comment
Share on other sites

Since we touched that topic of dark nebulae rendition in images, I would also like to add that I'm not quite convinced that true color those objects is brown/reddish.

I would expect it to be more neutral gray (for some reason I associate it with Moon dust rather than rusting metal). Quick search on google on this particular target yields quite diverse results - some of which indeed have such neutral gray coloring of dark nebula, and one explanation of red/brown appearance of such nebula might be due LP/color balance issues - most deep sky imagers contend with red gradients and background in their images when doing OSC. When doing mono and filters it's easy to remove background by histogram manipulation but care must be taken when doing color calibration (or white balance) as there could be residual red left due to LP in lowest values that would render other wise gray areas - reddish.

What do you think?

Link to comment
Share on other sites

7 minutes ago, wimvb said:

As others have said: great image.

Of course I had to do just that ?

There seems to be a satellite trail in the blue master. If you have PixInsight, you could try large scale pixel rejection. This way you don't need to increase the rejection parameter in pixel rejection (and lose valuable data), and still get rid of this trail.

I thought that too until I saw the matching one at 90 deg..   

Link to comment
Share on other sites

3 minutes ago, ollypenrice said:

Stunning image but, rather than just chuck out a subjective value judgement like that, I'd like to be specific.

Vlaiv's point is a serious one. However, we have the technology to make the dust a 'region of interest' and to find within it variations in illumination and subtleties of structure which, surely, are worth exploring. OK, so we do this at the expense of its 'true' relationship with the more-or-less vacuum of the background sky but why not? How interesting is that background? I prefer Peter's dust. The point of this image is to lift it into visibility. Also, it takes no inversions of the histogram to render the dust brighter than the background. It is brighter. Simple as that. But the point of stretching is, of course, to exaggerate that extra brightness.

If anyone but Peter had posted 'Iris in RGB' I'd have been twitching to say, 'No no no, for the dusty stuff you need loads of luminance.' How Peter manages without it in this way, I have no idea. I'm agreeing with Sara here, I think. Well, hats off, Peter. I need yonks of luminance and you don't!

And those elusive pinks are coming through - in an unreasonably short total exposure time. Humph!!!

Respect.

?lly

You have explained far better than I could of Olly and I'm pretty sure you could get amazing results of the dusty stuff with just RGB....

I know that this is basic information for you Olly but for the benefit of others I feel an explanation of my method might help a little.  

I do all my RGB binned at 1x1 so not to compromise the resolution. My camera has 9un pixels so each pixel has a much larger surface area than most allowing to capture more photon hits. That combined with my fast F3.8 Newtonian and bortle 4-3 skies gives me a fighting chance. A good set of flats and Bias frames are a must but I don't do dark's as my camera does not have set point cooling, I've found that my dark frames don't match the lights and when ever I've applied them they tend to do more damage than good. So I rely on dithered exposures and sigma clipping stacking methods to help with the noise. 

Although the data set is only RGB which I combine as you would normally. I do also combine all of the image data together to create a pseudo Luminance layer which I process as I would any Luminance. This then allows me to put my RGB layer though a heavier noise reduction without compromising the image quality. 

  Teasing out the dark stuff from the background with lots of smaller iterations is key. Measuring carefully the brightness of the areas of interest against the darker by means of the curves tool in Photoshop. When pushing the process down to the noise levels  the finer gradients really show up so I tend to push the saturation further in order to exaggerate them. I then set about correcting these gradients (which I do manually), then I bring back the saturation to a more reasonable level. .........easier said than done though it can take a few attempts to get it right.

Link to comment
Share on other sites

30 minutes ago, wimvb said:

As others have said: great image.

Of course I had to do just that ?

There seems to be a satellite trail in the blue master. If you have PixInsight, you could try large scale pixel rejection. This way you don't need to increase the rejection parameter in pixel rejection (and lose valuable data), and still get rid of this trail.

 

24 minutes ago, Laurin Dave said:

I thought that too until I saw the matching one at 90 deg..   

you are not wrong they are there .....and a real pain ... unfortunately I dont have enough frames for them to vanish completely 

n7023 pre process with sats.jpg

Link to comment
Share on other sites

3 minutes ago, peter shah said:

 

you are not wrong they are there .....and a real pain ... unfortunately I dont have enough frames for them to vanish completely 

n7023 pre process with sats.jpg

Firstly a great explanation of your processing principles.

Regarding trails, AstroArt, my stacking programme, has an outstanding 'remove line' filter which, if necessary, I'll use on individual subs to remove sat trails before combing in sigma clip. They may not vanish entirely but, once diminished, will generally vanish in stacking. Highly recommended.

There is one other trick if you have Noel's Actions for Ps. If you have a residual trail, rotate the image so this trail is vertical, then make a copy layer and run the 'remove vertical banding' routine on it. Go to the top layer and erase just the bit with the trail, flatten and restore the original image orientation.

Olly

Link to comment
Share on other sites

Archived

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

  • 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.