My first narrowband image - what am I doing wrong? - Forums

Topic starter

Mar 8, 2021

Having gained some experience with OSC imaging, I tried my first narrowband image last night.

NGC3579 (Statue of Liberty) was my target. I took 25 x 300sec frames in each of Ha, SII and OIII.

I must say that I am pretty disappointed with the result. And I am not sure whether it was the conditions (4-5arcsec seeing, thin cloud throughout0, my observations (should I have taken more SII, or longer exposures?) or my processing skills.

The images were processed in PI using the Weighted Batch Preprocessing script (which has worked really well for me on OSC images),
SHO-AIP to merge the resultant masters (equal weight to each master) using the standard Hubble palette, followed by the using post-processing workflow (at least for me) of Background Correction, MLT noise reduction/histogram Transformation/TGVDenoise/LocalHIstrogramEqualisation/StarReduction/Curves Transformation on the "luminance" part - not on the saturation (it was garish enough already) and a hint of Unsharpmask at the end.

The result is uninspiring. What an I doing wrong? Or it is just the data. Perhaps I should have chosen something with a little more SII.

With thanks in advance for the advice.

Brian

Helpful Respectful Engaging

John

Mar 8, 2021

nope ,nothing wrong. if you do the scnr tool,it will remove the green and makes it nice blue.

but give this a try in pixelmath
S2 in red
S2 +O3 in green
O3 in blue
Ha luminance with the LRGB tool. just trow the l layer over the above rgb image with s2 and o3.

greetings john

Brian Boyle

Topic starter

Mar 8, 2021 · Edited: Mar 8, 2021

John
Thank you so much for that brilliant advice. SCNR (why didn't I think to use it - its been a long time) works to give a lovely blue and orange image. I also tried you other recipe, it worked but did produce quite as pleasing an image (but better than the green). But I will continue to experiment. PixelMath is a great suggestion.
Thanks so much
Brian

David Redwine

Mar 8, 2021 · Edited: Mar 8, 2021

There is fix, if the green pixels are brighter than the red pixel replace the red pixel value with the green pixel value. This is usually done using Pixel Math in Pixinsight.

https://thecoldestnights.com/2020/06/pixinsight-dynamic-narrowband-combinations-with-pixelmath/
You can do almost the same thing in PS:
1. Split the color channels in PS, discard the L channel
2. Select the red layer and paste it over the green layer, set the blend mode to lighten
3. Recombine as follows:
     Red= the R+G channel
     Green= the R layer
     Blue = the B layer

The results:

Helpful

Tim Hutchison

Mar 8, 2021

Hi Brian:

If you're looking for a comprehensive narrow band example, you might want to check out the tutorial that Connor Matherne did for Deep Sky West. Here is the link:
https://www.deepskywest.com/tutorials

Eric Coles did one as well that I'm sure is also great. I only worked through Connor's. If you don't know these guys, both are very talented, accomplished APers. I'm sure you will find them helpful.

Best.
Tim.

Well Written Concise Supportive

Brian Boyle

Topic starter

Mar 9, 2021

Thanks Tim, I had a look at the link. Those tutotrials look perfect. Although a little expensive, with your recommendation I think it is a worthwhile investment. CS Brian

Tim Hutchison

Mar 9, 2021

Brian Boyle:
Thanks Tim, I had a look at the link. Those tutotrials look perfect. Although a little expensive, with your recommendation I think it is a worthwhile investment. CS Brian

Hi Brian:

I certainly understand WRT the price, but I worked through Connor's tutorial and I certainly feel like I got my money back in improved processes and techniques I did not know. If you haven't looked at his stuff here on AB, you certainly should @Connor Matherne . Like I said, he's very talented.

@Eric Coles (coles44) worked on this exact object in his tutorial. Maybe that hits a little closer to home for you. Admittedly, I did not work through Eric's yet, but he is also a terrific astrophotographer with WAY WAY WAY more experience that I have, and I'm sure you can't go wrong there either.

If you have any questions before you buy, I'm sure Lloyd from @Deep Sky West (Lloyd) would be happy to answer them. He's a really great guy and offers terrific customer service.

No matter which you choose (or both?) let me know your thoughts on it. I would like to hear what you think.

Best.
Tim.

Helpful Respectful Engaging Supportive

Brian Boyle

Topic starter

Mar 9, 2021

Thanks to everyone for advice.

Based on a compilation of the advice I received (SCNR, Hue correction and the various links), here is my latest effort.

Less Kermit and more Liberty in the Sky with Diamonds.....

Jerry Yesavage

Mar 10, 2021

Hi,

You are making rapid progress. Not sure you have seen this tutorial about color mapping:

https://thecoldestnights.com/2020/06/pixinsight-dynamic-narrowband-combinations-with-pixelmath/

Jerry Yesavage

Mar 10, 2021

Also, a great way to cheat on this is a program called 3D LUT that can work all sorts of magic on a Hubble pallet image… but it is pricey. It will cut out most of the steps in the prior tutorial.

Tayson

Mar 10, 2021 · Edited: Mar 10, 2021

All files stretch in pix in ez processing - soft stretch - default option.

Try to remove strars by starnet, go to SHO-AIP and: SII - Red 100%, Ha - Green 50-70%, OIII - Blue 100%.
In API select first upper option …rescali.

Now You may use some trick - autocolor in photoshop

(not always necessary)

Do not fight with stars, add it from ha - without color.
when You will get good result on nebula them stars will be next step.

Related discussions

WBPP Integration creating star artefacts (non symmetric) - or am I wrong?

I need some help with this. A "standard" WBPP processing of the M42 has created (to me) strange artefacts on the biggest stars in the Master Light file: Facts: Askar FRA400 och ASI2600MC, 80 x 60 sec exp. 30 Flats and 30 Dark Flats. To me i...

WBPP processing artifacts may explain disappointing narrowband results from integration.

Feb 5, 2023

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...

Narrowband integration time recipes directly relevant to author's Ha/SII/OIII approach.

Nov 6, 2022