Back to forums

Help with processing NGC 6543 core detail in PixInsight

Pleiades Astrophoto PixInsight Processing Help Needed NGC6543 NGC Objects Planetary Nebula
12 replies•312 views
Roger Renberg avatar
Roger Renberg avatar

Hi everyone,

I’m working on processing NGC 6543 (Cat’s Eye Nebula), but I’m struggling to bring out the detail in the core, and the OIII gets blacked out after adjusting the levels. I can see the structure in my masters, so I believe the data itself is solid, but I don’t have the experience in PixInsight yet to reveal it properly.

Does anyone have advice on effective techniques for handling objects like this in PixInsight? Are there any good tutorials or lectures specifically focused on processing bright planetary nebulae with high dynamic range?

I’ll leave some screenshots of the masters for reference.

Thanks a lot in advance!

📷 HaScreenshot 2025-09-11 at 21.09.44.png📷 OIIIScreenshot 2025-09-11 at 21.09.57.png📷 SIIScreenshot 2025-09-11 at 21.10.07.png

Well Written Respectful Engaging
Iosif Bodnariu avatar
Iosif Bodnariu avatar
Hi.

You should have sent the original images directly from the stack. What you sent cannot be combined. Put the original images somewhere so they can be copied and processed. Thanks.
Brian Puhl avatar
Brian Puhl avatar

Before I would even touch that data, I’d make an attempt to get it properly calibrated. Something don’t look right. From the thumbnail point of view I see a mild grid pattern, when zoomed in I see some splotchiness that I don’t believe is representative of the region.

andrea tasselli avatar
andrea tasselli avatar
I don't think you have much in there but hard to judge with the auto-stretch and I'm pretty sure you overexposed the core too so it is going to be a though job there. OSCs wouldn't have this issue though.  To give a helpful advice I'd need to see the stacked masters but in principle you need to use multiple stretching functions and isolate the core with a brightness mask. Blending needs to be done in non-linear space too. I'm assuming PI here, too. Cue: AutoHistogram and CurveTransformation.
Habib Sekha avatar
Habib Sekha avatar

I processed my NGC 6543 RGB data in PI using GHS (GeneralizedHyperbolicStretch). GHS requires some practice. There are various YouTube videos out there.

My NB 300s exposure data were overexposed, 60s were better. It is very easy to overexpose the core.

Here is mine using GHS in PI:

https://app.astrobin.com/i/xeko8d

Roger Renberg avatar
Roger Renberg avatar

Iosif Bodnariu · Sep 11, 2025 at 07:40 PM

Hi.

You should have sent the original images directly from the stack. What you sent cannot be combined. Put the original images somewhere so they can be copied and processed. Thanks.

Hi all,

First of all, thank you for the reply. The original files can be downloaded here https://we.tl/t-7S9XwZYweq

Habib Sekha avatar
Habib Sekha avatar

Nice data Roger! The nebula looks really fantastic, however, the core is overexposed.

Even without stretching the core is visible as a thick white circle.

Much shorter exposure length could fix this. There might be other solutions which I’m not aware of.

Well Written Respectful Concise
andrea tasselli avatar
andrea tasselli avatar
It's overexposed, with poor seeing and too low an image scale. And wrong telescope.
Erik Westermann avatar
Erik Westermann avatar

As the others have said, the core is pretty bright.

You could protect the brightest area using a mask. There are various ways to make a mask, but you could try the GAME or FAME scripts

Tony Gondola avatar
Tony Gondola avatar

The core of this object is one of the brighter ones out there, if not the brightest. I suspect that in your case the data just isn’t there and even if it was, at that image scale you’re not going to see much. I honestly don’t think there is any saving it in software.

Well Written
Jan Erik Vallestad avatar
Jan Erik Vallestad avatar

The Oiii channel is completely blown out, Ha is blown out a bit but the Sii channel looks just fine IMO. As always we should keep an eye on the histogram while imaging to avoid this. If it’s blown out in the RAW linear image without any preview stretching then it’s gone, if not then it’s all good.

Luckily though, you still had details in both Sii and Ha that could be used to work around the issues with Oiii for the core at least. The region between the core and the outer parts of the nebula seem more washed out. Gradients and off-looking star color aside, this is how it came out:

📷 DoneV2.jpgDoneV2.jpg📷 Done.jpgDone.jpg
I’ve not used anything but masked stretching/HT stretching, but a fair bit of layer work in PS was included as well. I’ve used variously stretched images of each channel to blend into one master, where the core has been preserved as much as possible. Then I also created a synthetic luminance layer from those.

The lower image has the synthetic lum applied to the core area, otherwise the images are alike. Also keep in mind that software like BXT can make the issue worse.

EDIT: Adding core view at 250%
📷 image.pngimage.png

Helpful Supportive
Roger Renberg avatar
Roger Renberg avatar

Oh wow,

Thanks for this! I’ll try to replicate the steps you’ve taken. I guess I’ll also need another set with shorter exposure times for the Hα and O III channels.

Well Written Respectful
Jan Erik Vallestad avatar
Jan Erik Vallestad avatar

Roger Renberg · Sep 14, 2025, 05:45 AM

Oh wow,

Thanks for this! I’ll try to replicate the steps you’ve taken. I guess I’ll also need another set with shorter exposure times for the Hα and O III channels.

Yes, that would be beneficial. I’d still do this approach myself in order to maintain a good looking core area while getting the fainter outer layers visible too.

Just do gradual stretches to a various degree on several copies, then mask/layer/blend them together. Masked stretch in PI works well initially, otherwise I prefer histogram transformation myself. Selective masking and a paint brush in PS is a simple and intuitive way of doing the blending process to get each master the way you want it.

In the end I think I had about three to four images that I blended for the Sii channel, three for the Ha channel and two for Oiii.

You don’t really need the synthetic luminance as the core naturally blends when combinging the desired palette in PI (I used pixelmath). Especially with the formulas I decided to use. But you can get a bit more fine details in the core by adding it carefully there.

Helpful