NGC 2264 - The Christmas Tree Cluster

[chris_g]

hi Chris,

Here y'go. I tried in PS first but I'm so rusty in it I was not getting anywhere. So this is in PI. The key truned out to be doing deconvolution and noise reduction before doing Starxterminator - still in linear. then stretch the starless, normal colour, and contrast adjustments then stretching the stars minimally before adding back in. Giving this;



I'm sure it could be refined more - for example stretching the R & B channels seperately then recombining as HOO (since this is l-enhance date) and some final noise reduction may help. maybe seperate out the Ha and put it in as a Lum layer.

The only problem with the data is really the stars. They are quite bloated and many are saturated which surprises me when you have 180s, with your 600d and the L-enhance. What ISO did you use.

Anyway, i hope this is of some value to you.

take care

David.

Figure all that out but one thing, what is "still in linear" ?

[Jockinireland]

hi Chris,

Here y'go. I tried in PS first but I'm so rusty in it I was not getting anywhere. So this is in PI. The key truned out to be doing deconvolution and noise reduction before doing Starxterminator - still in linear. then stretch the starless, normal colour, and contrast adjustments then stretching the stars minimally before adding back in. Giving this;



I'm sure it could be refined more - for example stretching the R & B channels seperately then recombining as HOO (since this is l-enhance date) and some final noise reduction may help. maybe seperate out the Ha and put it in as a Lum layer.

The only problem with the data is really the stars. They are quite bloated and many are saturated which surprises me when you have 180s, with your 600d and the L-enhance. What ISO did you use.

Anyway, i hope this is of some value to you.

take care

David.

Figure all that out but one thing, what is "still in linear" ?

Hi Chris,

"Linear" refers to the data before it has been stretched in any way - so before levels, curves etc. Once it has been stretched it becomes "non-linear" and thats when it starts to become visible to our eyes.

Astro specific processing apps like PI, Startools and APP allow us to make a temporary stretch which is only on screen and not on the actual data and then allows us to do certain processes, for example gradient removal, deconvolution and noise reduction on the linear data. This, for reasons too mathematical for me to explain or understand, is considered (by some people) to be a much better way to process AP data.

In PS that becomes much more difficult if not impossible because there is no way to work on the data in linear form - you cant see anything. I dont know how well things like gradientXterminator, Carbonis noise reduction and StarXterminator would work if you did them on an image which was stretched JUST enough to see whats going on.

Take care

David.

[chris_g]

hi Chris,

Here y'go. I tried in PS first but I'm so rusty in it I was not getting anywhere. So this is in PI. The key truned out to be doing deconvolution and noise reduction before doing Starxterminator - still in linear. then stretch the starless, normal colour, and contrast adjustments then stretching the stars minimally before adding back in. Giving this;



I'm sure it could be refined more - for example stretching the R & B channels seperately then recombining as HOO (since this is l-enhance date) and some final noise reduction may help. maybe seperate out the Ha and put it in as a Lum layer.

The only problem with the data is really the stars. They are quite bloated and many are saturated which surprises me when you have 180s, with your 600d and the L-enhance. What ISO did you use.

Anyway, i hope this is of some value to you.

take care

David.

Figure all that out but one thing, what is "still in linear" ?

Hi Chris,

"Linear" refers to the data before it has been stretched in any way - so before levels, curves etc. Once it has been stretched it becomes "non-linear" and thats when it starts to become visible to our eyes.

Astro specific processing apps like PI, Startools and APP allow us to make a temporary stretch which is only on screen and not on the actual data and then allows us to do certain processes, for example gradient removal, deconvolution and noise reduction on the linear data. This, for reasons too mathematical for me to explain or understand, is considered (by some people) to be a much better way to process AP data.

In PS that becomes much more difficult if not impossible because there is no way to work on the data in linear form - you cant see anything. I dont know how well things like gradientXterminator, Carbonis noise reduction and StarXterminator would work if you did them on an image which was stretched JUST enough to see whats going on.

Take care

David.

I was wondering if that was what you meant, I tried running Star Exterminator after running just gradient removal processes, I use AstroFlat Pro and HLVG, the nebula was pretty much gone. I also found a couple of actions that do a deconvolution using Camera Raw. I know it's there, my skills aren't just quite there to pull it out looking any kind of decent, yet. Also, all the data is L-Pro, not L-eNhance

[BABOafrica]

The image without the nebula is quite striking! I like it a lot.

The image with the nebula is good. Hard to process those...

BABO

[jthommes]

Cool Chris. It's interesting to see the star luster with the nebulosity not catching the eye.

[STEVE333]

Hi Chris - I saw the beautiful job David did on enhancing your data, so, thought I would give it a try. It took me several tries to achieve the image below. I used Starnet2 to remove the stars while the data was still linear. I like the way it works. I also used EZ Star Reduction to reduce the number and brightness of the stars. Hope you like it.

Cheers,

Steve

[STEVE333]

Here's a version where the stars have been reduced even further and the background made a little darker.

Steve

[chris_g]

Here's a version where the stars have been reduced even further and the background made a little darker.





Steve

Hi Steve,

I'm curious about how much you stretch an image. My skills at processing are still way in their infancy. I just figured out that stretching something and keeping something linear are two different things. Stretching is just making the image bigger, that's easy enough to do in Photoshop. StarNet sets PI's auto stretch is usually sufficient. Do you have any idea how much that is? Also, whenever I don't do any curves at all and just use gradient removal tools, the nebula data is pretty much gone. Not just this image, any image I've tried it on. So that makes me wonder, exactly what is stretching and how does it differ from keeping things linear?

[STEVE333]

Hi Steve,

I'm curious about how much you stretch an image. My skills at processing are still way in their infancy. I just figured out that stretching something and keeping something linear are two different things. Stretching is just making the image bigger, that's easy enough to do in Photoshop. StarNet sets PI's auto stretch is usually sufficient. Do you have any idea how much that is? Also, whenever I don't do any curves at all and just use gradient removal tools, the nebula data is pretty much gone. Not just this image, any image I've tried it on. So that makes me wonder, exactly what is stretching and how does it differ from keeping things linear?

Hi Chris - "Stretching" an image doesn't mean making an image bigger. It refers to using something like Curves in Photoshop where the low level signals are boosted more than the high level signals. This is done to bring out details in faint signals while trying to avoid saturating stars or other high level signals. Before any "stretching" the data (image) is considered to be linear. Once it is "stretched" it is considered to be non-linear. Hope this makes sense.

Steve