Dual Band Filter OSC, - Forums

Topic starter

Jun 12, 2025 · Edited: Jun 12, 2025

Not sure if I can call this processing or imaging. Last night I tried my hand at OSC Dual Band filtering. I used an optolon L-Extreme.

Hours data on a 2 different targets. I have now tried 10 different processing methods.

All have the stars looking nice and round and also colored, (WAY TOO COLORED) See included image.

I and run star extraction and my nebula looks good, and have the correct data but the stars are making me crazy. Any ideas what I am doing wrong?

newstars.png

This is from PI, I stretched it, ran StarXterminator and these are the stars I get. I have no idea what I am doing wrong. Any suggestions?

Habib Sekha

Jun 12, 2025

Run StarExterminator before stretching. Stretch the stars separately according to your taste. then at the end , add the two images together using Pixelmath. HTH

Clayton Ostler

Topic starter

Jun 12, 2025 · Edited: Jun 12, 2025

Habib Sekha:
Run StarExterminator before stretching. Stretch the stars separately according to your taste. then at the end , add the two images together using Pixelmath. HTH

I tried that already the stars are still the wrong colors

If I just open the FITS file and use SFT to look at it all the stars are still messed up. I am out of ideas

Noah Tingey

Jun 13, 2025

You're looking at narrowband stars. That's just what they look like. Either live with them or take separate RGB stars and add those in.

Mike H - Sky View Observatory

Jun 13, 2025 · Edited: Jun 13, 2025

There's also a halo around the bright star. I sold my L Extreme because of it and moved to a L Ultimate which doesn't have the same halo effect.

Mike

David Foust

Jun 13, 2025 · Edited: Jun 13, 2025

Try the NB to RGB stars tool that @Franklin Marek developed! It really helps correct dual NB star color:

https://www.setiastro.com/pjsr-scripts

He's also got a ton of other useful tools. I use the statistical stretch, star stretch, automatic DBE, and FAME tools pretty regularly as part of my processing routine.

His HaloBGone tool might help reduce the appearance of some of the star halos too.

Helpful Supportive

Clayton Ostler

Topic starter

Jun 13, 2025

Thanks. This is my first round with a dual band filter and I just assumed that stars would work like they do on a broadband filter. This is my mistake. It seems like what I have is kind of expected I just got to figure out how to process it out

David Foust

Jun 13, 2025

Yeah the star colors don't look right straight out of a dual NB image. The script helps significantly. It's not perfect but much much better. The other alternative is that you could shoot an hours worth of 60 second frames with just a UV/IR cut filter and combine those stars with your narrowband nebulosity.

Helpful

NeilM

Jun 13, 2025

After NG to RGB i use SPCC, followed by StarX.

Clayton Ostler

Topic starter

Jun 13, 2025

After NG to RGB i use SPCC, followed by StarX.

Id love to learn more about your work flow order. I'm wondering if I am doing his out of order making things worse

Right now I.

Calibrate and Stack
Crop
Noisex
Image Solve
SPCC
Blurx
Graxpert
DBXtract into OIII and HA
From each of those I StarXtermintor
Then I use ForaxPallet Utility to put Ha and OIII back together
Then use NG to RGB to out my stars together
The I use pixel math to blend the 2 images together

Still stuck will rainbow stars. And my background is present terrible with chroma noise.

Love this hobby

Miguel T.

Jun 13, 2025

Take a look at the posterization effect in the gradients. You're not working in 16 bits, or at least not exporting/previewing in 16 bits like you should. I'm not sure if that even is 8 bits. It look like the 256 color palette from the 90s. Fix that first.

andrea tasselli

Frequent contributor

Jun 13, 2025

Enable 24 bit LUT

Robin Bosshard

Jun 13, 2025

Not directly related to the star's colors, but the workflow:
"Noise reduction of any kind should never be applied before deconvolution of any kind, including BlurXTerminator"
(source: rc-astro)

I do BlurXT right after SPCC like you but denoise later on…

CS!

Eric Gagne

Jun 13, 2025

Clayton Ostler:
After NG to RGB i use SPCC, followed by StarX.

Id love to learn more about your work flow order. I'm wondering if I am doing his out of order making things worse

Right now I.

Calibrate and Stack
Crop
Noisex
Image Solve
SPCC
Blurx
Graxpert
DBXtract into OIII and HA
From each of those I StarXtermintor
Then I use ForaxPallet Utility to put Ha and OIII back together
Then use NG to RGB to out my stars together
The I use pixel math to blend the 2 images together

Still stuck will rainbow stars. And my background is present terrible with chroma noise.

Love this hobby

Like others have said, if you want nice stars, shoot them in broadband and process them separately. I also think you're not helping yourself by running NXT before SPCC and that you are running background extraction way too late in your workflow. IMO it should be the first step.

My typical workflow without RGB stars is.
Background extraction
BXT (Correct only) Russel Croman recommends doing this before SPCC.
SPCC
BXT
SXT
NXT
SCNR (if wanted/need)
Stretch to your heart's content
Stretch the stars. I use SetiAstro's star stretch
Put starless and stars images together. I use Screenstars from Mike Cranfield & Bill Blanshan
Most of the time I will reduce the stars with StarReduction, also from Mike Cranfield & Bill Blanshan

You might want to try it. I do not claim this to be the ultimate workflow and your mileage will undoubtedly vary but maybe some of it will help.

Helpful

Clayton Ostler

Topic starter

Jun 13, 2025

Eric Gagne:
Clayton Ostler:
After NG to RGB i use SPCC, followed by StarX.

Id love to learn more about your work flow order. I'm wondering if I am doing his out of order making things worse

Right now I.

Calibrate and Stack
Crop
Noisex
Image Solve
SPCC
Blurx
Graxpert
DBXtract into OIII and HA
From each of those I StarXtermintor
Then I use ForaxPallet Utility to put Ha and OIII back together
Then use NG to RGB to out my stars together
The I use pixel math to blend the 2 images together

Still stuck will rainbow stars. And my background is present terrible with chroma noise.

Love this hobby

Like others have said, if you want nice stars, shoot them in broadband and process them separately. I also think you're not helping yourself by running NXT before SPCC and that you are running background extraction way too late in your workflow. IMO it should be the first step.

My typical workflow without RGB stars is.
Background extraction
BXT (Correct only) Russel Croman recommends doing this before SPCC.
SPCC
BXT
SXT
NXT
SCNR (if wanted/need)
Stretch to your heart's content
Stretch the stars. I use SetiAstro's star stretch
Put starless and stars images together. I use Screenstars from Mike Cranfield & Bill Blanshan
Most of the time I will reduce the stars with StarReduction, also from Mike Cranfield & Bill Blanshan

You might want to try it. I do not claim this to be the ultimate workflow and your mileage will undoubtedly vary but maybe some of it will help.

Thanks for the tips and advice. Will try and see what I get.

Clayton Ostler

Topic starter

Jun 13, 2025

Miguel T.:
Take a look at the posterization effect in the gradients. You're not working in 16 bits, or at least not exporting/previewing in 16 bits like you should. I'm not sure if that even is 8 bits. It look like the 256 color palette from the 90s. Fix that first.

I agree. How do I change that though? My fits is the same in PS, and PI, and ZWO Windows Tools.

The ASI2600mc is supposed to be 14bit camera.

andrea tasselli

Frequent contributor

Jun 13, 2025

Clayton Ostler:
The ASI2600mc is supposed to be 14bit camera.

The ASI2600MC *IS* a 16-bit camera.

Clayton Ostler

Topic starter

Jun 13, 2025

andrea tasselli:
Clayton Ostler:
The ASI2600mc is supposed to be 14bit camera.

The ASI2600MC *IS* a 16-bit camera.

Regardless, those images are not looking even 12bit

andrea tasselli

Frequent contributor

Jun 13, 2025

Just send over the xisf/fit file(s) and we shall see what's in there.

NeilM

Jun 13, 2025

Hi Clayton

looks like you're getting a lot of feedback which is great!

For me, I use RGB to NB earlier in the process than you do. Basically I have one process for the stars and then another for the HOO/SHO image, and then I combine them at the end.

For the stars, I do Gradient Correction, NB to RGB, SPCC, BXT, StarX, then I stretch the stars and usually use curves transformation to make sure the color saturation is to my liking, I discard the starless image because the NB to RGB process messes up the non-stellar stuff.

For the starless, I start over using with gradient correction, BXT, DBExtract, etc.. Obviously don't use NB to RGB on the starless image.

I will add that my usual "process" doesn't always work to my satisfaction on every image - it depends on the target. If something doesn't look right I'll change things up.

Lastly, I do sometimes use RGB stars (shot without a filter) instead of the NB to RGB ones. I have even used stars from old images that I took before I had the NB filter. I have found the "Star Alignment" process to be fantastic in allowing me to use stars shot with different framing and even different focal lengths, in order to align them perfectly with the NB image.

Good luck!
Neil

Helpful Engaging

Rick Veregin

Jun 14, 2025

I have the same camera and the same filter, I have never seen anything like this, no matter how it is processed--the weird look has nothing to do with the filter or camera. What does the image look like if you open the raw fits it in ASIfitsview for example? It should have a greenish background and stars will likely look white, as this simple stretch in that program doesn't add any saturation or color balance. Here is a typical example, I right clicked on the image and chose histogram from the menu, you can see I have a 16-bit file.

If it is not 16-bit, then you need to find the right settings in your acquisiton program, there is a bit depth setting. Also, make sure you are saving as a fits file in acquisition. There are also typically all sorts of settings in acquisition to change bit depth, color balance, etc, etc. Make sure everything is set back to defaults. What software are you using to get your images?

Hope this helps
Rick

Helpful Supportive

Clayton Ostler

Topic starter

Jun 14, 2025

I have a long way to go, but thank you for those who kicked in ideas

Heres where I landed for this

Stack/Calibrate
Crop
DBXtract
For each of the 2 extract StarXterminate
Graxpert on each of the 2 OII and HA
Forrx Pallet Contruction (not adding stars just the OIII and HA)
Blur xTerminator
Noise Externimator
Seti Astro create RGB Stars
Stretch new Forax RGB
Image Blend

I landed here https://astrob.in/xqb49l/0/

Its not amazing but at least the stars better.

kgav8r

Jun 23, 2025

I'm a little late here, but I personally prefer the Foraxx Palette utility stars over the Seti Astro. It seems to give a bit more color variety. However, the script has not been updated for the latest PI release. You can install it manually, though. HERE IS A THREAD with the download link and a high level description of how to install.

Well Written Concise

Related discussions

Problems with stacking a Galaxy in Pixinsight

Hey guys, i am still a total noob regarding pixinsight. last night I imaged the pinwheel galaxy for a few hours with my duo narrowband filter. the individual frames looked good but the stacked picture I get out of pixinsight is absolutely unusable. I...

Feb 22, 2023

Both posts describe amateur astronomers who captured seemingly good individual frames with narrowband filters but encountered problems during the processing/stacking stage that made their final images unusable.

Blending RGB OSC with L-Enhance data - ZWO 294MC Pro

Hey Everyone! After trying to find a forum topic to answer my question, I was not able to properly get the answer I am looking for. First of all, I image with a refractor telescope using a 294MC Pro. I usually only use no filters at all in Bortle5 or...

Sep 6, 2023

Both posts describe astronomers experimenting with different imaging equipment and techniques but appear to be cut off mid-sentence before completing their thoughts.

Confusing Results - OSC - Adding Ha to RGB data.

I decided that M82 (Cigar Galaxy) would be a great target to try combining RGB data from a OSC camera with DNB data (Also OSC) but the results I got have left me confused. I assumed the best way to image this target would be to shoot the galaxy with ...

Jan 7, 2024

Both posts describe astronomers struggling with processing techniques for one-shot color (OSC) camera data and encountering unsatisfactory or confusing results.

How far can you go with Narrowband Integration times arround an hour? Share your recipes and results!

To give you a little background on this post let me share what I am trying to do: One of the reasons for buying a wide lens (Askar ACL200) was to be able to play with composition for my Astro images in more widefield pictures first. I have started to...

Nov 6, 2022

Both posts are incomplete, cutting off mid-sentence while discussing astrophotography techniques and equipment.