I think my images are going well until I try to color them in the color module

[xonefs]

Non-linear stretching and then combining makes no sense... Is it likely you just stretched until you saw the detail "you liked" (which is mostly visible in the Ha) and then decided it was time to combine?

That's what it is supposed to do and what its purpose is, but squashing and stretching the dynamic range, detail and the per-pixel SNR prior, doesn't make a whole lot of sense.

To be more precise I did combine them first before stretching and then extracting luminance. Though I did try the other way too and have in the past. Yes, I fully admit this is a bastardized workflow and goes against best practices. I put that aside for a moment anyway to go with what I was familiar with learning from other people who also were probably teaching poor practices from my limited experience just to compare.

Then I am frankly not sure what you are after any more... It shows all the detail you were after that you indicated in this post and is better(!) visible than in your PI/ST hybrid on all my screens. I am really wondering if there is something amiss with your screen calibration? Do you have, say, a phone or tablet handy? What do the images look like on those?

Yeah I'm sorry if this is confusing- it is kind of hard to describe and be precise with what is actually jumping out at me. A lot of this is just overall big picture initial reaction and hard to pinpoint what I'm seeing since there are many different things going on that all contribute to the overall look and feel of an image. Looking at it zoomed out gives an initial impression different from now when I am zooming in at those particular features where they look better. While zoomed out I think I am seeing some of the effects of the HDR module creating a certain texture/look over parts of the image that is what I am having a reaction to and am not a big fan of.

Zooming in it does look like there is decent detail on those specific areas and looks better. There is then overall depth and contrast of the nebula to the background, and around the entire "rim". I thought my hybrid one appeared the least flat overall, for whatever reasons and specific factors are contributing to my perception of that.

Maybe my perception would change if I saw it in color, as that gives another form of contrast.

My monitor is not calibrated, but I have looked on a few. Time is also a factor as I notice different things viewing at different times and if I have been staring at any of them too long.

Now I'm looking at yours again bigger on its own screen and I'm starting to like it more, but looking at it zoomed out and smaller when it's in a post less so. Visual perception is strange.

It does not add noise depending on the band (unless a fixed amount of noise is present - thermal or shot noise from skyflow, etc.). This is the main reason for why you would make a linear synthetic luminance frame, weighted according to exposure times . And it's also the reason why stretching channels individually and then mixing them into a luminance frame yields wildly varying noise signatures.

ok that's interesting, I think I am starting to understand that.

Not enough in terms of detail? Color? Were the objects poor in Ha signal/detail?

this was a while ago and memories from prior processing methods (where I probably learned and was emulating bad techniques from others). I don't think I tried using just an Ha lum and had noisy SII and Oiii that was becoming problematic in terms of noise in the image and background.

I have not tried it much yet in startools really with narrowband

I have noticed processing LRGB images that it seemed like I had too much luminance and not enough data on RGB channels to color them well. I would try to shoot 2-3x as much luminance as individual RGB times to try to get a more clean luminance, and it often appears faint stuff like dust or edges of galaxies wasn't colored well.

[xonefs]

I don't want to change course too much but your comments on methods made me try processing again in ST on an image I had issue with before and thought I just didn't have enough data

It's about 2.5 hrs Ha and a little over 3 hrs Oiii. The oiii still seems pretty weak

I first tried adding the oiii and ha to luminance with their exact times and processing (I left one at Oiii at 0 since it was loaded into both green and blue). I seemed to get a lot of noise in background this way.

So I tried processing again using just Ha for the luminance and Oiii for color only and had better results.

This is not the soul nebula of my dreams but overall I think it looks good and details look nice for the data quality (though prob needs more saturation looking on a different monitor). I am not having any of the reactions I was with all the lagoon attempts, so I think it is something about that target that is more challenging.

data is here: https://drive.google.com/drive/folders/ ... sp=sharing

I am curious how people get so much blue from Oiii in the middle. I had assumed I would have to shoot a lot more time on Oiii. or is that just very selective processing people do? In the other active thread on soul nebula here and the OP seemed to get a lot of blue without much time.

Edit: very curious about this now, as my oiii looks better than some others I've seen get blue color in the middle of their processed SHO or HOO image with similar or less.

[Mike in Rancho]

Thinking of one possibility about the "look" that you are after.

Whether intentional through some kind of selective manipulation, or perhaps just the way it works, a lot of PI images can have sharply defined edges around some details, such as nebulous clouds you circled, but a lot of the rest of the object is soft, fuzzed, out, even blurry. I see this constantly. And it can in fact be pretty, and often gives a 3D effect to those sharper cloud edges. I think that look is in your sample links, as well as the ones you have posted that you seem to like better. It's also been all over the place lately in a lot of the Pacman images (current CN target).

This may also explain why you want to throttle back HDR so much.

I am not sure how they are doing it, but I don't believe ST is going to discriminate that way for you -- enhancing only certain detail whilst leaving adjacent detail soft, unless that can occur through the various detail size and scaling settings in contrast or HDR.

All I can think of is to work your controls in contrast in HDR, and perhaps also sharpen settings in order to get those edges (which are probably highlight, not shadow?).

Maybe? That's the only thing I can really come up with as to the "overall feel" differences, which you are in fact correct about. Though I don't know if it's proper processing.

Perhaps Ivo can explain, if that is the image look that you are seeing.

[xonefs]

^mike I do think I know what you’re describing and have noticed that in the past- it is from using luminosity masks with sharpening and denoise (you can see that on my rosette). I don’t like that so much which is a big plus for ST and I didn’t employ any of that in my hybrid one (ez denoise might do that under the hood but not to any obvious level I can notice) and don’t really notice that on some others as being the factor.

[happy-kat]

whoops wrong thread

[Mike in Rancho]

^mike I do think I know what you’re describing and have noticed that in the past- it is from using luminosity masks with sharpening and denoise (you can see that on my rosette). I don’t like that so much which is a big plus for ST and I didn’t employ any of that in my hybrid one (ez denoise might do that under the hood but not to any obvious level I can notice) and don’t really notice that on some others as being the factor.

Darn. Thought I was on to something.

Maybe do a side-by-side, perhaps of zoomed-in areas, like Ivo often does for comparisons, to get across the difference you are looking for. Though indeed if it is the feel of the whole image, that'll be tough.

Thanks for the luminosity mask explanation though! Things are starting to come together for my understanding of what I see in all the PI images that get posted.

[admin]

very curious about this now, as my oiii looks better than some others I've seen get blue color in the middle of their processed SHO or HOO image with similar or less.

You are definitely on the right track - if O-III is unusably noisy for whatever reason, then just use Ha for luminance but keep O-III for color (e.g. set green/blue total exposure to 0 in the Compose module).

As always, all you need to do in the Color module, is to reduce red (or increase gren and/or blue) to make O-III pop through. It's that dead simple;

StarTools_2806.jpg (378.03 KiB) Viewed 6279 times

or

StarTools_2807.jpg (372.37 KiB) Viewed 6278 times

...etc.

Just remember that, in StarTools, Color (ideally) not dependent on detail and the other way around. You can boost the coloring of the O-III to your heart's content.

As for things looking different on different scales, it's totally a thing!

My personal rule of thumb is that it has to look "good" at 100% zoom; the fine detail ("high frequency detail") is where it counts, and is what we spend the most time optimizing our images for; perfect focus, perfect tracking, waiting for perfect seeing, applying deconvolution, etc. all to get the fine details as perfect as we can. Might as well make them the star of the show.

[Burly]

The trouble with colour perception is there’s so many You tubers kicking out images showing blue colouring done in pixinsight that people think that that’s correct though may look pretty isn’t correct colour for oiii.

[xonefs]

You are definitely on the right track - if O-III is unusably noisy for whatever reason, then just use Ha for luminance but keep O-III for color (e.g. set green/blue total exposure to 0 in the Compose module).

As always, all you need to do in the Color module, is to reduce red (or increase gren and/or blue) to make O-III pop through. It's that dead simple

Ok thanks- those look very good!

To clarify when you say set to 0 in compose module should they bet set to the lowest number which is 0hr01min or should they read "not set"?

I am trying to replicate your settings in color and am not getting anything near that, which is what I was running into before when I had tried to do that in color:

like the closest:

It doesn't work any better in the legacy mode. is this a problem from wipe or something?

edit:

I think there is something else going on. I re-wiped and bumped dark anomaly filter to 2 and it looked like it changed it and I was able to get more color separation but still not looking quite like that

is this how wipe should look here?

[Mike in Rancho]

Interesting. I was able to get decent-ish separation even using DAF 1 on it, NB preset. Though I didn't slide the OIII exposure times to zero, I just currently tried it using the times noted -- I think 9300 and 12000? There is a fair bit of blue noise in the background. DAF 0 showed that in Wipe big time, so I set it back to 1. 2 didn't seem all that different. I suppose I can try the Ha only for L option and OIII just for color.

I presume you are setting it to bicolor in Compose?