[RCC] Western Veil Nebula - Forums

Topic starter

May 25, 2026

Hi AstroBin Community,

I am looking for areas on where I can improve my image I took of the Western Veil Nebula in late August last year over 5 nights. I used my ZWO ASI183MC Pro, Radian 61 (with its original objective lens), Optolong L-Ultimate and Sky-Watcher Star Adventurer GTi.

This is one image I have been struggling to process for quite some time, especially considering I want the background Ha filaments to stand out just enough without overpowering the OIII structures. I tried using Photoshop, Lightroom and PixInsight to produce something I am happy with, but I cannot seem to find a happy medium.

One of the biggest issues I am having is both StarNet2 and SXT removing the finer filaments. SXT leaves really clean backgrounds whilst removing nebulosity whereas StarNet2 keeps much more filaments but leaves a messier background which I have to clean up in post.

Attached is my most recent edit of this data. I don’t think I’ve done a good job with this one (as has been the case for many months) as I notice it’s quite “mushy” and soft whilst having a somewhat magenta-ish tint to the Ha filaments. The saturation seems washed out as well from what I can tell.

Looking forward to receiving feedback for this image so I can take it further than before!

Zak

📷 Western Veil Nebula.jpg

Well written Respectful Engaging

Dave Stirling

May 26, 2026

What does your processing flow look like? I’m sure folks can dive in with some suggestions after seeing approximately what you’re doing.

I am still relatively new at PixInsight but found I was struggling with gradients and other activities in the linear phase of processing that had serious downstream effects after I stretched. The “messier background” might be a symptom of gradient processing, or due to over-aggressive stretching. I’ve found that Generalized Hyberbolic Stretch is much easier for me to keep my “overstretch” trigger finger in check.

Well written Helpful Respectful Concise Engaging Supportive

Zak Jones

Topic starter

May 26, 2026

I generally try and keep the same workflow for all my images regardless of type e.g. HOO, SHO etc. I don’t know if this is the “right” way to approach workflows for every single image I process lol.

My typical workflow is usually structured to something like this below. It’s different for my SHO and HOO images with my ZWO OSC cameras as this Western Veil Nebula image was taken before I got my QHY miniCAM8 late last year. I also tend to try and tame my stretches on my images regardless of stretching method to prevent them from appearing too aggressive or “fake” with too much contrast or colour saturation.

Hopefully this example workflow of mine can help give some pointers where I might be making mistakes!

OSC HOO

Gradient Removal (e.g. MultiscaleGradientCorrection/GradientCorrection, Seti Astro Automatic DBE or GraXpert)

BXT Correct Only

SPCC with Photon Flux

BXT Sharpen

DBXtract

ChannelCombination for HOO

SXT or StarNet2 on Both HOO and Ha Images

Starless Images

NarrowbandNormalization on HOO Image

NXT or DeepSNR on Both Images

Stretch Both Images (e.g. GHS, Seti Astro Statistical Stretch, MultiscaleAdaptiveStretch or EZ Soft Stretch)

LHE, UnsharpMask, Curves, SelectiveColorCorrection, MultiscaleMedianTransform, ColorSaturation, RangeSelection on Both Images

ImageBlend Script to Blend Ha Image onto HOO Image

HOO Stars Image

NarrowbandNormalization (Mode 3, 0.300 Blend Amount)

BXT Correct Only

Seti Astro Star Stretch

ScreenStars Script

Final Small Adjustments (e.g. Colour, Noise Reduction etc.)

ICCProfileTransformation to Convert ProPhoto RGB Images to SRGB for Web and AdobeRGB for Print

F11 on SRGB Image to Balance Colours and Black Levels

Export Processed Image

Dave Stirling

May 26, 2026

You’ve got a lot of tools in play for your workflow. I can only give you a few suggestions to poke at the edges because I’m not tracking the rationale for every step you have — all of which may be justified based on your shooting conditions and what you ultimately want from the product.

Working on the original OSC image without a channelextract at all, just to see where the symptoms show up, might be illuminating. You might also try a GHS → Linear stretch after the GHS → GHS to take your left edge in the histogram to the low clip point and see how your filaments look vs. the noise.

This step: “LHE, UnsharpMask, Curves, SelectiveColorCorrection, MultiscaleMedianTransform, ColorSaturation, RangeSelection on Both Images” may be a meta I’m not aware of, but seems formulaic/busy to me, would be good to understand what you’re solving for in this step that couldn’t be LHE + Curves, for instance. In my own painful experience, I rarely get anywhere solving for a flaw in the image by adding more tools and steps.

You shouldn’t need NBN on your star image, I’d just fiddle with curves to get the aesthetic you want.

That’s all I’ve got, only been at this for a few months and learning as I go, so take the free advice for what it’s worth. ;)

Well written Helpful Insightful Respectful Engaging