[RCC] What do you think of my 20m Eastern Veil image?

15 replies763 views
AmyWarble avatar

https://app.astrobin.com/i/v3vvaj

My procedure is in the image description, reproduced at the end of this post.

I previously captured this nebula using 5 minute exposures. I was interested in seeing what a much higher exposure would do, so I captured this using 20 minute exposures. I was amazed at the detail I captured, but I’m disappointed to only receive 11 likes on the image.

So I wanted to ask: what do you think of this image? What are its pain points? What should I do to excel?

————————————————

For each SHO master:

  • GraXpert

  • BlurXTerminator

  • NoiseXTerminator

  • StarXTerminator

NB to RGB Star Combination + Star Stretch -> Stars
Combine SHO starless -> Nebula

On Nebula:

  • GHS stretch

  • AutoLinearFit

  • NBNormalization

  • NoiseXTerminator

  • UnsharpMask

ImageBlend Stars + Nebula

Helpful Engaging
SemiPro avatar

The downfall of most Veil images is that it’s extremely tempting to over-sharpen the filaments. I would start by toning down your sharpening. Usually there is the added problem of a poor SNR that cannot accommodate the processing. Fortunately for you, your data does not seem to have this issue!

Maybe try some traditional DBE and see what that gets you. I am not exactly the foremost expert of what the kids are using these days, but AI-based DBE had a nasty habit for nuking faint details and was only really useful for images that had poor initial calibration. Keep in mind the last time I seriously processed an image was around April 2024.

https://www.astrobin.com/yq4ahe/?no-redirect

Check that out for an idea of what level of detail you should be going for. I figure you might be able to press it just beyond that, but its always such a delicate balancing act!

I also think you will have trouble attracting viewers with the colour palette you chose. What can I say, people like the red!

📷 image.pngimage.pngLooks like something happened here as well, with the green line.

Helpful Insightful Respectful Engaging Supportive
Brian Puhl avatar

This image is way over sharpened. It honestly kinda hurts to look at. Really need to back off on the BlurX on filamentary structures like this. You took it even farther by using unsharp mask, or another round of blurX after stretching. It’s just wayyy too much. Less is better in astrophotography. I can overlook the colors, but they’re not my taste. The blackpoint seems to be neutral, so everything else about the image is fine to me.


BlurX is an amazing tool, but it must be used in moderation. Unsharp Mask has it’s place, but very rarely, and only an extremely low power.

Also, @SemiPro That streak in that screenshot is a real structure. It’s been talked about before.

Helpful
AmyWarble avatar

Thanks for the feedback everyone!

Brian Puhl avatar

If you attempt a reprocess, you are welcome to continue posting in here and get continued feedback. I may be blunt, but I am 100% willing to help.

Well Written Respectful Supportive
Matthew Singer avatar

It’s a great image with lots of nice detail.

I think the main issue is the green throughout; there are several ways to address it.

Well Written
Tony Gondola avatar

I really like the diffuse glow along the top but I don’t like the color choices or the garish saturation. There’s lots of detail but as others have mentioned, it’s over sharpened. When working with sharpening you really want a light hand. Apply the effect and then go back and back it off until it’s just a little boost. A little goes a long way.

Well Written Concise
AmyWarble avatar

How could I address the green? I can adjust the SCNR during NarrowbandNormalization, but that erases it. I could change the color palette?

I uploaded revision B just now. I turned the settings on BlurXterminator down, and didn’t do any additional sharpening.

Matthew Singer avatar

NarrowbandNormalization is very helpful because it can be applied to both linear and non-linear images, and can be used multiple times to make gentle adjustments.

  • NBN’s SCNR is not the same as PixInsight’s SCNR - it behaves differently (see the comments below from Bill Blanshan’s original PixelMath expression that ultimately led to the NBN process)

  • Experiment with NBN’s lightness options both in linear and non-linear

  • Try to gently adjust the colors using NBN when the image is linear as well as non-linear

As with everything PixInsight, experimentation is key, and using processes in slightly different ways can have significant effects.

Here’s Bill’s comment from his original PixelMath expressions:

This pixelmath equation was written for a few reasons;

1) The current Pixinsight SCNR process uses an image

“Lightness” to correct the image appearance after

removing the green color cast. “Lightness” works

within the LAB color space and should not be used on

linear Images! This pixelmath process can work on

linear and non-linear images. :-)

2) The current Pixinsight SCNR process can also clip

Red or Blue channels when removing green color casts.

This pixelmath process does not allow for clipping :-)

Helpful Insightful Respectful
AmyWarble avatar

https://app.astrobin.com/i/v3vvaj?r=C

I gave it another go. This is HOO with SII added. Sharpened stars, nebula was barely sharpened.

Lynn K avatar

BlurXterminator in not a shapenting tool. It is AI deconvolution and as Russ Croman has stated several time on line, it should only be use on linear data or can cause artifacts.

Lynn K.

AmyWarble avatar

Lynn K · Sep 27, 2025, 03:56 PM

BlurXterminator in not a shapenting tool. It is AI deconvolution and as Russ Croman has stated several time on line, it should only be use on linear data or can cause artifacts.

Lynn K.

Okay? I’m only using it on linear data and I’m not using it to sharpen.

SemiPro avatar

AmyWarble · Sep 27, 2025, 02:34 AM

https://app.astrobin.com/i/v3vvaj?r=C

I gave it another go. This is HOO with SII added. Sharpened stars, nebula was barely sharpened.

It’s looking much better! I would be tempted to see what just a tad bit more sharpening on the nebula looks like, but the red haze/filaments on the outer edges are pretty much perfect in my opinion.

Well Written Respectful Engaging Supportive
Brian Puhl avatar

I much prefer this version, the image doesn’t feel forced, or heavy handed. The signal seems well balanced and the background is neutral. The luminosity is great. You might have room to up the deconvolution power a bit with BlurX, but less is always better in my book. My only critique is the color, or lack there of. I feel a little less green and a bit more saturation. That aspect is an artists taste though and does not demerit the overall quality of the image IMO. I think you did very well and fixed every point that was addressed.

Well Written Helpful Insightful Respectful Engaging Supportive
Jared Holloway avatar

Looks great!

Alex Nicholas avatar

Overall brilliant… I would love to see it processed as HOO with no SII, Just red and blue to show the distinct differences in the signal between the two.