Final Masters

• LHaRGB: stars, starless
• H-alpha: stars, starless
• Luminance: stars, starless

Description

Here’s M31, the Andromeda Galaxy, our nearest major galactic neighbor, sitting about 2.5 million light years away. This mosaic took 3 panels to capture, combining H-alpha, Luminance, and RGB data for a total of 76 hours and 47 minutes of integration.

Let me know what you think, and as always: clear skies!

Close Ups

Outer region - left The center of M31, accompanied by M32 Outer region - right Messier 110

Final Masters

H-alpha

O-III

S-II

1. Intial Post: H-Alpha

Some work in progress on the Cygnus Wall (NGC 7000). You might recognize this from an earlier RedCat51 shot of the North America Nebula.

Data was captured remotely in Romania under beautiful Bortle 3 skies. 5h50m of H-alpha frames, each 300s. Same setup I use at home: ASI533MM Pro and Askar 71F.

It’s amazing to see the motion in the gas and dust streams here. If you look closely, you can spot stars forming inside the dark structures, especially in the darker region just above the center right.

H-alpha Data

2. Update: Oxygen-III added

Some more progress on the Cygnus Wall (NGC 7000). This is an HOO palette image, containing both Hydrogen-alpha (Ha) and Oxygen-3 (O-iii) data. Mapping HOO to RGB channels gives it this colors.

Still waiting for the last batch of Sulfur-2 data to complete the final image but wanted to share it nonetheless.

H-alpha and Oxygen-III

3. Final addition: Sulfur-II

Another collab coming to an end! This is the final version of the Cygnus Wall (NGC 7000) in SHO-palette, containing Sulfur-2, Hydrogen-alpha (Ha) and Oxygen-3 (O-iii) data. Mapping SHO to RGB channels gives it this colors.

Final result in SHO

🔭 Clear skies!

Final Masters

H-alpha

O-III

Crop of the Nebula

Crop

Description

This is the Crescent Nebula (NGC 6888), located in the constellation Cygnus.

In its center lies a massive Wolf-Rayet star, which is shedding its outer layers through intense stellar winds. These fast winds slam into slower material that the star expelled earlier on, creating shockwaves that form the surrounding gas into a glowing ‘cosmic bubble’, the puff you can see here.

It’s an emission nebula, consisting mostly of hydrogen and oxygen. Therefor I chose to create a HOO palette image.

🔭 Clear skies!

Final Masters

Luminance

H-alpha

Red

Green

Blue

Processed Luminance

Processed luminance

Description

I present to you my very first dark sky image: Messier 101, the Pinwheel Galaxy ✨️ — captured under Bortle class 3 skies using remote observatory data in collaboration with JoepsAstronomy. This project totals a whopping 25 hours of integration, combining broadband (LRGB) and narrowband (Hα) to reveal the galaxy’s sweeping spiral arms and faint outer halo.

The pink patches scattered across M101’s arms are Hα regions—massive clouds of hydrogen where new stars are born. Even more amazing: while M101 lies roughly 21 million light-years away in Ursa Major, all the bright foreground stars in this image belong to our own Milky Way. Two galaxies, one frame.

This was a true processing challenge, balancing the rich LRGB data with the more intense Hα signal, done in PixInsight. I focused on enhancing the fine structures without over-emphasizing noise or losing the galaxy’s natural colors. Especially the core needed some final tweaking to make the dust swirls stand out more since it is pretty bright.

Captured remotely with the same setup as my own: ASI533MM Pro and the small but powerful Askar 71F. Each frame was 300 seconds long, collected over multiple nights.

🔭 Clear skies!

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. SpectrophotometricFluxCalibration,
2. DynamicBackgroundExtraction,
3. PixelMath to combine the RGB data with the Hα: $T * 0.2 + Ha * 0.8 (yep, should have used median instead),
4. ImageIdentifier
5. SpectrophotometricColorCalibration,
6. BlurXTerminator,
7. MaskedStretch for stretching the image while keeping the bright parts of the galaxy under control,
8. HistogramTransformation for final stretches
9. NoiseXTerminator on the image using settings: HF: 0.95, LF: 0.70, Scale: 7.4, Iterations: 5,
10. DynamicBackgroundExtraction to remove the background gradients,
11. CurvesTransformation for color correction,
12. ChannelCombination to merge the Luminance data with the HaRGB data,
13. UnsharpMask for sharpening the image a bit

Final Masters

Synthetic Luminance

Ha

SHO Combined

RGB Combined

Description

I present to you, the heart of the Heart Nebula: Melotte 15 ❤️. After weeks of cloudy skies, I finally had four clear nights to collect data for this image! This deep exposure combines narrowband data (SHO) with broadband (RGB) star data. The final integration time adds up to 17h 20m, making this my longest project in a while!

Processing was done in PixInsight, focusing on balancing the faint structures while keeping a natural starfield. The RGB stars were captured separately on the last night to prevent stars from becoming purple in the calibration process.

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. SpectrophotometricFluxCalibration,
2. DynamicBackgroundExtraction,
3. BlurXTerminator,
4. SpectrophotometricColorCalibration,
5. ColorCalibration on Melotte 15 to match the RGB channels levels,
6. ChannelCombination for merging a SYnthethic Luminance with RGB data,
7. MaskedStretch for stretching and keeping the bright parts of the nebula under control,
8. StarXTerminator for removing stars from the SHO data,
9. Final stretch with HistogramTransformation,
10. Creation of a so called ‘synthetic luminance’ by combining the SHO data with SNR as the weighting factor. The synthetic luminance was then combined as the CIELab L channel on the SHO data.
11. NoiseXTerminator on the image using denoise: 1.00:0.91 (color:luminance), low frequency denoise: 1.00:0.80 (color:luminance), frequency scale: 6.6, and 5 iterations.
12. Used NarrowbandNormalization to balance the SHO data.
13. Color correction with CurvesTransformation on the background image
14. Sharpening with UnsharpMask on the background image
15. Noise reduction with NoiseXTerminator on the background image
16. Adding the RGB stars on top of the SHO data with PixelMath to prevent purple stars.
17. FastRotation for better angle :)
18. ImageSolver for adding the astrometry data to the image.

Final Result

Please check the image in full resolution on my Astrobin post.

Final Masters

Synthetic Luminance

RGB Combined

Description

You might think you recognize this image. And you’re absolutely right!

I’ve posted an image with the same object, the Pleiades (M45), back in September 2022 on Instagram, just when I started my astrophotography journey. Since we’re dealing with rain again and the Pleiades season is upcoming, I decided to put my new Pixinsight workflow to the test on this older dataset.

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. Weighted Batch PreProcessing,
2. DynamicBackgroundExtraction,
3. ImageSolver,
4. SpectrometricColorCalibration,
5. BlurXTerminator,
6. HistogramTransformation,
7. ChannelCombination for merging a SYnthethic Luminance with RGB data,
8. StarXterminator,
9. CurvesTransformation on the background image
10. NoiseXTerminator on the image. 0.8:0.23 (Denoise:Detail) applied to the background and 0.64:0.30 to the reflection nebulae using masks.
11. PixelMath for merging the starless image with the stars.

Final Result

Please check the image in full resolution on my Astrobin post.

Final Masters

O

S

H

Description

So here it is… I present to you: my tri-band image of the ✨ Eastern Veil Nebula ✨!

This weekend, I captured the final set of data using the Sulfur (S-II) filter, completing my final image.

Astrometry Calibration

HSO, SHO, HOO, OSH?

When using narrowband filters, which isolate specific wavelengths of light, you create images similar to Hubble’s famous ‘Pillars of Creation.’ Typically, the filters are assigned as follows: Sulfur (S-II) to red, Hydrogen (H-alpha) to green, and Oxygen (O-III) to blue—also known as the Hubble Palette (SHO).

When shooting dual-band filters, like Hα and OIII, you can create a different palette. The mose common palette in this case is the HOO palette, where Hydrogen (Hα) is assigned to red, Oxygen (OIII) to green, and Oxygen (OIII) to blue. If I’d only applied two channels of data (dual-band) the image would look like the following:

You often see images of the Veil Nebula in this blue-green and red palette, looking like an old-school 3D glasses picture. However, depending on how the data is applied, the results can vary. For this image, I aimed for a more ‘realistic’ look, similar to how you might perceive it with the naked eye. I found that the OSH palette (Oxygen, Sulfur, Hydrogen) worked best for me. Of course, it all comes down to personal preference!

Noise reduction (new approach)

I’ve been working on noise reduction techniques for most of my images and found out there is a better way on reducing noise on narrowband images. With LRGB data you could use the Lumminance channel for noise reduction. However, with narrowband data, you don’t have a Lumminance channel.

For this problem you could create a so-called synthetic lumminance channel by combining the three narrowband channels. This can be created in Pixinsight using the ImageIntegration tool. Just add the three masters (S, H, O) as Input Images and set the weights for the Image Integration setting to SNR.

Hit the Apply Global (cirkel) button and you’re done. You now created an image with the best signal-to-noise ratio of the three channels combined. This image can now be applied as lumminance channel on the narrowband image by merging it on CIE L a b mode using ChannelCombination.

Pixinsight, ChannelCombination settings

As you can see, by default, the image both contains less noise and more detail!

Before - After applying the SNR image as lumminance channel

SNR Image

Below is the final result for the SNR image:

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. FastRotation on the OSH image (180 degrees),
2. DynamicBackgroundExtraction on the OSH image,
3. ImageSolver for preparing the image for correction in SPCC,
4. SPCC,
5. BlurXTerminator on the OSH image with values: 0.5 sharpening stars, -0.1 halo adjustment and 0.7 non-stellar sharpening
6. Stretched the image using HistogramTransformation
7. Prepared an OSH SNR image using ImageIntegration and ChannelCombination as described above and applied it as lumminance channel
8. StarXTerminator (splitting stars and background),
9. Adjustments with CurvesTransformation
10. NoiseXTerminator on the starless image using masks for separate background and galaxy noise reduction
11. Star intensity control and (size) reduction using only the brightness curve in the CurvesTransformation tool. Also pumped up the saturation a bit.
12. Merging of the stars and starless image using PixelMath, expression: $T + OSH_STARS
13. Final contrast adjustments with CurvesTransformation

Final Result

Please check the image in full resolution on my Astrobin post.

Final Masters

Ha

R

G

B

Description

This is 🐴 IC434 🐴 mostly known by its other name: the Horsehead Nebula.

Ever seen those three almost perfectly aligned stars in the night sky during winter? The lowest star (seen from the northern hemisphere) is surrounded by a massive hydrogen cloud where stars can be formed and has a diameter of 1500 light years.

Shot this image using a Hydrogen-alpha (Hα) filter and regular red, green and blue filters.

Merging H-alpha with RGB data

This image needed a different approach. Since IC434 contains lots of Hydrogen, I shot with a Hydrogen filter rather than a Lumminance filter. This way I could filter out the details of the nebulae. After combining red, green and blue channel data, normally, you’d merge it with luminance data. Instead, I merged the hydrogen data with the red channel using PixelMath. For this, the following statement was provided on the red channel:

# PixelMath
B: $T
G: $T
R: $T * 0.05 + Hα * 0.95

Below you can see the H-alpha master file I used

H-alpha Inverted

For most of the narrowband images I create an inverted image as well. This way details are more visible than normal. Below is the inverted Hα-master.

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. DynamicBackgroundExtraction on separate channels,
2. LRGBCombination on only RGB channels
3. SPCC,
4. Merging Ha with Red channel on RGB image using expression: $T * 0.05 + Ha * 0.95
5. Rerunning SPCC again for exact spectral match of stars and therefor enhancing red channel,
6. BlurXTerminator on the HaRGB-channel for correcting stars and nebulae
7. SNCR noise reduction (0,70) to reduce green noise,
8. BackgroundNeutralization to make sure background is correctly calibrated
9. Stretched the image using HistogramTransformation
10. DarkStructureEnhance on the image (0,40) to enhance horsehead nebula
11. StarXTerminator (splitting stars and background),
12. Adjustments with CurvesTransformation
13. NoiseXTerminator on the starless image using masks for separate background and galaxy noise reduction
14. Merging of the stars and starless image using PixelMath, expression: $T + HaRGB_stars
15. Final adjustments with CurvesTransformation
16. BlurXTerminator for reducing star size a bit

Final Masters

L, 300s

L, HDR

Red

Green

Blue

Update: 28 Dec 23

Removing flat frames

After almost a year I’ve updated the M42 image with another revision of the same data. I ditched the flat frames since these where taken with a way too short exposure time creating bands of read noise. The ideal exposure time for flat frames is 3+ sec for astro cameras.

Adding High Dynamic Range (HDR)

I’d taken 30 sec exposures as well so I decided to try and test out the HDR workflow. This includes a combination of both HDRComposition (for merging the two luminance images masters to a high dynamic range image) and the HDRMultiscaleTransform.

A great workflow of creating HDR images in PixInsight can be seen on the video below:

PixInsight Wokflows: instructions on how to process images as HDR

Description

Another run on M42 after a full strip down of the HEQ5 Pro and applying some thorough maintenance. Shot this new footage of M42 together with some new flat frames since my master flat frame had some banding issues. Luckily this could be fixed easily by making new ones with longer exposures.

Maybe time for an Hα-filter? 🙃

The image below shows a high contrast visual of the annotated M42.

M42 Inversed, high contrast

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. Weighted Batch PreProcessing,
2. DynamicBackgroundExtraction,
3. ChannelCombination,
4. ImageSolver,
5. PhotometricColorCalibration with background neutralization,
6. BackgroundNeutralization,
7. HistogramTransformation,
8. Masked appliance of NoiseXTerminator and BlurXTerminator.
9. Final adjustments (LRGB merging color and saturation) done in in Photoshop.

Final Masters

Luminance

Red

Green

Blue

Description

Normally, I am very critical of my work, but I am secretly quite proud of this result.

The weather is nothing but cloudy these days. Yesterday we’ve had one night of clear sky and M31 in Zenit. Guess what the target was…

This is my third attempt to capture M31, our closest neighboring galaxy. Its distance from us is “only” 2.5 million light years, which is relatively close in astronomical terms. Wanted to shoot H-alpha channel as well but trees were blocking the view by then. Might add those later.

M110, Andromeda’s satellite galaxy

M110 Galaxy

Another thing to keep in mind during the processing of this image was to make sure enough details of M110 remained visible. As you look closer, some dust cloud are visible.

Astrometry Calibration

Data Processing

Processing was done in PixInsight only. Steps of the process are listed below:

1. DynamicBackgroundExtraction on separate channels,
2. LRGBCombination on only RGB channels
3. SPCC,
4. SNCR noise reduction (0,70) to reduce green noise,
5. ArcsinhStretch a couple of times to stretch the RGB image where colors remain having their calibrated color value,
6. BlurXTerminator on the L-channel for correcting stars and nebulae
7. LRGBCombination to merge the noise L-image on the RGB-image
8. StarXTerminator (splitting stars and background),
9. Color adjustments on the starless image
10. NoiseXTerminator on the starless image using masks for separate background and galaxy noise reduction
11. Merging of the stars and starless image using PixelMath, expression: $T + STARS
12. DarkStructureEnhance on the image (0,40) to enhance galaxy dust bands
13. Final adjustments with CurvesTransformation
14. Last small bit of noise reduction using NoiseXTerminator
15. Star reduction/resizing using the StarDe-emphasis script (add the url https://pixinsight-updates.astroswell.com/ to your repositories)