Jump to content

2052839955_Mobilephonebanner.jpg.a502a319d7033354d442937f2edd0c2c.jpg

IKO - M81 & M82 LRGB+Ha - Processing Competition


Grant
 Share

Recommended Posts

16 minutes ago, gorann said:

It is not only us and the IKO data that swings that star towards blue, I just saw this post by Skarpen on Astrobin:

https://www.astrobin.com/w399qv/

But as Wim points out - our main goal is to make pretty images😁

I think that there is general tendency to shift images to blue part of spectrum.

Most of the stars are simply yellow or orange. Such images look both dull and also they look "imbalanced". I think that most people have issue with star color because when they see daytime image that is taken in warm light - like candle light or 2700K incandescent light and overall tone is still yellow / orange - we automatically think that image needs to be color balanced.

image.png.39148c46697d076dd4522307b0d73b28.png

If you compare these two images - most of us will say that left is "properly" color balanced.

Our brain tends to do the same when sees predominantly yellow / orange stars in the image - we "instinctively" think that image should be sort of "neutral grey" (grey world hypothesis - basis for one type of auto white balance algorithm).

 

Link to comment
Share on other sites

Here's my effort- brought to you by lots of cloudy nights when I'm not able to gather my own data!

I've been on holiday and so had a bit of time to play around and thought I'd put together a video of my main steps- partly to invite feedback on how I processed it, and partly on the off-chance that it might give others ideas. The act of documenting it has been quite educational in itself- primarily in showing me that what in my head is a reasonably structured and methodical process is actually a bit chaotic and all over the place. Hey- ho- it's been fun.

If you've got a few minutes this is the video: 

And here's my effort:

708335056_GlowRemoval.thumb.jpg.bce23bdd37ce0050feb0b71b223f769d.jpg

I've attached a TIFF as well in case too much detail is lost in the JPEG.

 

Final.tif

  • Like 5
Link to comment
Share on other sites

2 hours ago, wimvb said:

Darned. I thought I was in time with my image. 🤔😉

So is this not a live thread? I thought it was, though it does say 2020 in Grant's post at the start. There just don't seem to be any old responses, though the year isn't stated on the posts, just the date. I think most people in this thread thought it was live. I'm rather confused, though I'm used to that...

Olly

Link to comment
Share on other sites

21 minutes ago, ollypenrice said:

So is this not a live thread? I thought it was, though it does say 2020 in Grant's post at the start. There just don't seem to be any old responses, though the year isn't stated on the posts, just the date. I think most people in this thread thought it was live. I'm rather confused, though I'm used to that...

Olly

It's a typo by @Grant, because the thread was started in august 2021. Hence my emojis

Edited by wimvb
Link to comment
Share on other sites

Bit late to the party with this one. Didn't have as much time on it as i would have liked, but i found it pretty difficult to process i must admit. 

APP used for Gradient Reduction, Siril used for Photometric Colour Calibration, then into PS for everything else. Colour stretched manually in PS using Arcsinh, and the RGB was used for the galaxy cores. 

Didn't use starnet for a change. instead i just went with old-school star reduction. 

M81 & M82 IKI.jpg

  • Like 5
Link to comment
Share on other sites

Edit: I'll leave my attempt but I've just realised I was too late in entering in time so it's just to look nice.

Super data, thank you for sharing.

Processed in StarTools 1.7.461MR6

I hope to redo this process as it is a little too strong for me.

First step was load the ha2 as linear and crop | bin| wipe | autodev with ROI | contrast | HDR | mask and wavelet sharpening | decon | colour | superstructure | shrink | denoise | saved as ha2.tiff

Using the Compose module I loaded the R G B and L,  and when adding the luminance weighted it 20 mins.

RGB L process  loaded as linear and crop | bin | wipe | autodev with ROI | contrast | HDR |  mask and wavelet sharpening | decon | colour | | superstructure | shrink | denoise | saved RGBL.tiff

Compose module and opened saved ha tiff on red and as RGB legacy and no to colour interpolation.

Loaded as none linear then layer module and opened RGB and set to lighten (or might have been add) this created the final composition | saved tiff to png for upload.

The ha data has the warmth for the galaxy and also the orange for the star mentioned previously on this thread that showed blue though the halo after the ha addition stayed blue.

1147903173_NewComposite30082021endv1.thumb.png.5419054d9fa750b26638b10f175581d8.png

Edited by happy-kat
  • Like 1
Link to comment
Share on other sites

  • 2 weeks later...
On 15/09/2021 at 20:37, Viktiste said:

Any chance you could share some more details of your PI workflow? Wonderful result.

Thanks for your comment. I hope this helps.

 

 Referring back to my saved PI Project:

1.       Dynamic Crop and DBE on all masters.

2.       Renamed masters as Ha, Lum, Red, Green, Blue.

3.       Added Ha to Red using PixelMath, straight Red+Ha, rescaled, no mask. Renamed this as RedHa

4.       Added Ha to Lum using PixelMath, straight Lum+Ha, rescaled, no mask. Renamed this as LumHa

 

1.       Channel Combination using RedHa, G and B. Renamed this as RGBHa

2.       BN and CC on RGBHa.

3.       Masked Stretch on RGBHa.

4.       RGBWorkingSpace on RGBHa with all channels as 1.0.

5.       Extracted the Lum from RGBHa.

6.       Several small saturation boosts in Curves, using the extracted Lum as a mask.

7.       SCNR.

8.       HT to balance the channels.

9.       Convolution to slightly blur.

10.   Repeat RGBWorkingSpace on RGBHa with all channels as 1.0.

11.   Set this RGBHa aside for later.

 

1.       Noise Reduction on the LumHa using Jon Rista’s method (https://jonrista.com/the-astrophotographers-guide/pixinsights/effective-noise-reduction-part-1/).

2.       Stretch using several small iterations of HT until the galaxies are well defined, and then use several small iterations of Curves to bring up the background and IFN whilst keeping the galaxies controlled.

3.       HDRMT to compress the dynamic range of the galaxy cores, using 5 wavelet layers, de-ringing on, lum only on, no mask.

 

1.       If required, extract the Lum from RGBHa. Balance this Lum using LinearFit, with LumHa as the reference. Add this extracted Lum back into RGBHa using ChannelCombination, using CIE L*a*b*, with a* and b* disabled. This ensures that RGBHa is at a similar brightness as, and ready to receive, LumHa as luminance data. This step may not be necessary if both the luminance and colour are already at similar levels.

2.       Add LumHa into RGBHa with ChannelCombination, using CIE L*a*b*, with  a* and b* disabled.

3.      Make a star mask for the smaller stars and reduce their size/brightness using MT.

4.      Rename as LumHaRGBHa.

Edited by A320Flyer
Link to comment
Share on other sites

On 21/09/2021 at 12:39, A320Flyer said:

Thanks for your comment. I hope this helps.

Thank you so much for posting this. I have some follow up questions 😉

But perhaps the competition thread is not the correct place, so I made a new thread here

https://stargazerslounge.com/topic/383551-m81-m82-lrgbha-iko-dataset-processing-with-pixinsight/

 

Edited by Viktiste
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.