1.43
#...
·
 |
|---|
|
I've been refining my capture and processing technique, and just I've completed what I believe to be my highest quality image so far. Please review and give me any feedback you can offer about what and where I can still improve. This was fully processed and edited in PixInsight except for the JPEG export done by Photoshop (just because I think they do a slightly better job with details). https://astrob.in/c1lg6y/0/ I was aiming for vivid color here without going off the deep end of over-saturating. I would have liked the blue to be darker blue but I found that keeping it where it was preserved the fine detail better. Note the different shades of blue, and how this fades to green around the edges. My palette of choice lately for SHO images is a blend of "regular" SHO (just mapped to RGB, respectively) and the Forax style PiP-based SHO. This one is a 40/60 blend of those two. I also made the choice to set the background at about 10% and as neutral as I could make it. It seems like a good compromise between crushing the dark areas and leaving the image with too little contrast and pop. One note: I am still not 100% convinced I have the Flat 8 reducer on my FLT132 dialed in perfectly. I've never seen a reducer so sensitive to sub-millimeter back focus adjustments. The image here is slightly more cropped on the left vs the right (just for the sake of framing) but is still fairly close to the full frame (with the exception of stacking artifacts). Please let me know what you think of the stars in the corners. And thanks in advance for your time and feedback! |
1.20
#...
·
|
|---|
|
It has the natural processing feel that I've been trying for lately. You did not go off the deep end ! I also like the SHO/Forax palette. Your stars seem to be all white, without color. I've been trying different ways to put color in my stars but I think I will have to use the RGB filters to get it right. Overall, a beautiful image ! Monty |
|
1.43
#...
·
|
|---|
Thank you! I appreciate the feedback on the processing. I always conflicted between creating an image that pops and one that doesn't go too far.  Yes, the stars are mostly white although there are a handful of yellow/orange ones. Those are buried in the more orange parts of the nebula though, so they're hard to spot. There's a larger one in the middle near the top of the frame for example. These are SHO stars with the characteristic magenta removed. I need to shoot LRGB stars more often but I end up rarely doing so. Thanks again! |
 #...
·
·
1
like
|
|---|
|
Hi Greg! This image is (at least for me) a very well processed image, the right amount of denoise and sharpening. You stretched it fairly far as well, and why not, looks awesome. As the comments above, would be nice to get some RGB data (L not needed) into the image. I always like to compare images with mine to see the differences, in this case it is VERY different, but personally I like both of them. I applied just for fun Topaz on your image, of course without star removal, etc., just to see if one can get it a bit sharper, and indeed, I think with a little extra effort one could make that happen. Here the 2 comparison images, first one with mine, second one in Topaz.   You should be very happy with your image! Uwe |
|
1.43
#...
·
·
1
like
|
|---|
|
Thank you, Uwe! I appreciate the comments and feedback. I have had another look at the stars and I think I could pull more color into them using one of the tricks of the trade, R=H, G=0.x*H+0.(1-x)O, B=O, for some value of x. That's not RGB but does bring some more orange and blue tones in. It's a potential work-around for a case where I don't have the RGB data. Topaz may be the next step for me as I'm starting to feel decent about the rest of my processing. I need to read up more on it. I'm a little concerned about adding detail that isn't originally there (although frankly star removal does the same thing whether via StarNet or StarXterminator). What's your thought on the "AI" part of Topaz? I have to admit, that find detail you were able to tease out using Topaz is compelling. Wow! Also, your shot is beautiful! I love all the contrast and of course the detail in the Bok globules. Thanks again. |
#...
·
·
2
likes
|
|---|
|
Like the idea to get some Pseudo RGB stars added, I do that a lot as well. And of course, you asked the big question: how far is allowed in processing? Is Topaz still in the frame of being realistic? There were and are a lot of discussions about it, and I believe nobody can really answer it with certainty. I read the argument: if the process reduces deficiencies of your image caused by the camera and other things, well, then this is ok. But if it is adding extras, then it is not. I am not so sure if this is the final answer, although pretty straight forward. A LOT of images here on Astrobin, and especially images that got some kind of reward, use a lot of this extra. It starts already by pushing some of the channels much further than the others, this will give a false sense of the overall distribution of spectral colors. So, it really is a pretty made-up image. But an image that pops and is nice to look at! If I would be all scientific about it, I would discard all those images. But most of us are not that scientific, we want to see space as stunning, hence I do understand why this is done. So, all in all: it comes down to as with almost everything else: Everything in moderation cannot be bad! And .... if you like your image better that way, well ... this is what counts at the end!! Is it really important that others don't like your image if you do? |
|
1.43
#...
·
·
2
likes
|
|---|
|
Thanks for your thoughts on this. I agree, it's a controversial topic, like any denoise or deconvolution is as well, but there's nothing about astro imaging (especially narrowband) that's very realistic anyway. After you're done making the exposures, and *maybe* stacking, the rest of it is artistic license. Even removing gradients is a bit of mathematical fakery. I'm going for images that look beautiful to my eye, and I do enjoy when others like the result as well. We aren't doing photometry here (unless it's a processing step, which it often is). Thank you. I will look into Topaz. |
