Jump to content

Narrowband

M101 Processing Challange


HunterHarling

Recommended Posts

Thanks for sharing the data. I have always used a DSLR to date and last month bought my first mono astro imaging camera, so this gives me a great chance to practice post processing this type of data.

All done in Pixinsight and finished a little in PS.

 

 

M101-haRGB.jpg

Link to comment
Share on other sites

Thanks, I enjoyed that. Data is very good although there does seem to be some horizontal banding, probably caused by calibration files.

Processed in PS, PI and APP. I didn't bother doing a separate layer for the stars as i normally would. Just the usual sharpening, local contrast enhancement and high pass filter.

Thanks again.

Richard.

m101_Rgb_combine-St.thumb.jpg.4afff33540ecc0506b9255e693774cc1.jpg

Link to comment
Share on other sites

3 minutes ago, Allinthehead said:

Wow, amazing colour in that Wim. We're going to need a workflow i'm afraid?

Sure, I'll do a write up later. There was a lot of experimenation in this. Atm, I'm finishing up on the LHaRGB. "Stay tuned"

Link to comment
Share on other sites

1 hour ago, HunterHarling said:

Wow wow wow! This is amazing! How did you do it??

 

2 hours ago, Allinthehead said:

Wow, amazing colour in that Wim. We're going to need a workflow i'm afraid?

Here's the short version (if an image tells more than 1000 words, then 2 images should do an even better job)

Screendump_RGB.thumb.jpg.641fbeeec76b476fc532f5f7f9fe6ed7.jpg

Screendump_LHaRGB.thumb.jpg.38689cf3c041da111fd71d330fea7db7.jpg

The slightly longer version:

LRGB:

  1.  conversion of the masters to 32 bit floating point accuracy.
  2. RGB combination of the masters
  3. cropping the edges
  4. DBE to flatten the background as much as possible
  5. Canon banding reduction to get rid of vertical bands on the right hand side (read pattern, I suspect)
  6. Traditional colour calibration
  7. Noise reduction: TGVDenoise on L and MMT (8 layers) on Chrominance
  8. Masked stretch 3 times with 0 clipping up to a background level of 0.3 This jumpstarts colour saturation
  9. Finetuning the stretch (bringing in the blackpoint to just below clipping) and increasing colour saturation
  10. SCNR green
  11. HDR range compression and LHE on the midrange section, to get better definition of the galaxy
  12. Further tweaking of the blackpoint and a slight stretch
  13. Create a copy for Ha integration
  14. LRGB combination with a 50% mask, because I didn't like the full effect
  15. final adjustment of the blackpoint and highlights

The luminance was easier:

  1. Crop
  2. DBE
  3. Deconvolution
  4. Noise reduction (TGVDenoise)
  5. Histogram stretch
  6. HDR compression (default values)
  7. Blackpoint and highlight adjustment

Ha had a simple treatment:

  1. Crop
  2. DBE
  3. Canon banding reduction (vertical read pattern on the right hand side)
  4. Histogram stretch

I used the NBRGB script on non-linear images to combine the Ha with the RGB

Then did LHaRGB pretty much the same as for the RGB data

Ended with taking down the saturation of the stars over the galaxy a bit. I would like to revisit this step. There are still a bit of star halos over the galaxy, which I don't like.

Finally, resampled to 50 % (60 % for LHaRGB) and saved as jpeg with 97 % quality factor.

 

Link to comment
Share on other sites

5 hours ago, Allinthehead said:

 

hanks, I enjoyed that. Data is very good although there does seem to be some horizontal banding, probably caused by calibration files.

Processed in PS, PI and APP. I didn't bother doing a separate layer for the stars as i normally would. Just the usual sharpening, local contrast enhancement and high pass filter.

Thanks again.

Thumbs up on this one-nice job

Rodd

Link to comment
Share on other sites

10 hours ago, wimvb said:

 

Here's the short version (if an image tells more than 1000 words, then 2 images should do an even better job)

Screendump_RGB.thumb.jpg.641fbeeec76b476fc532f5f7f9fe6ed7.jpg

Screendump_LHaRGB.thumb.jpg.38689cf3c041da111fd71d330fea7db7.jpg

The slightly longer version:

LRGB:

  1.  conversion of the masters to 32 bit floating point accuracy.
  2. RGB combination of the masters
  3. cropping the edges
  4. DBE to flatten the background as much as possible
  5. Canon banding reduction to get rid of vertical bands on the right hand side (read pattern, I suspect)
  6. Traditional colour calibration
  7. Noise reduction: TGVDenoise on L and MMT (8 layers) on Chrominance
  8. Masked stretch 3 times with 0 clipping up to a background level of 0.3 This jumpstarts colour saturation
  9. Finetuning the stretch (bringing in the blackpoint to just below clipping) and increasing colour saturation
  10. SCNR green
  11. HDR range compression and LHE on the midrange section, to get better definition of the galaxy
  12. Further tweaking of the blackpoint and a slight stretch
  13. Create a copy for Ha integration
  14. LRGB combination with a 50% mask, because I didn't like the full effect
  15. final adjustment of the blackpoint and highlights

The luminance was easier:

  1. Crop
  2. DBE
  3. Deconvolution
  4. Noise reduction (TGVDenoise)
  5. Histogram stretch
  6. HDR compression (default values)
  7. Blackpoint and highlight adjustment

Ha had a simple treatment:

  1. Crop
  2. DBE
  3. Canon banding reduction (vertical read pattern on the right hand side)
  4. Histogram stretch

I used the NBRGB script on non-linear images to combine the Ha with the RGB

Then did LHaRGB pretty much the same as for the RGB data

Ended with taking down the saturation of the stars over the galaxy a bit. I would like to revisit this step. There are still a bit of star halos over the galaxy, which I don't like.

Finally, resampled to 50 % (60 % for LHaRGB) and saved as jpeg with 97 % quality factor.

 

Wow, thanks for this. I'm wondering, how did you got masked stretch to stretch it this much? Every time I use masked stretch it does not stretch it to my liking...

Link to comment
Share on other sites

Just now, HunterHarling said:

Wow, thanks for this. I'm wondering, how did you got masked stretch to stretch it this much? Every time I use masked stretch it does not stretch it to my liking...

I used it 3 times:

1. To background level 0.1, 50 iterations

2. To background level 0.2, 50 iterations, max background level increased from 0.05 to 0.15

3. To background level 0.3, 50 iterations, max background level increased to 0.25

All with clipping fraction set to 0

After that you bring in the blackpoint with histogram transformation.

You need a flat background for this to work.

Link to comment
Share on other sites

9 hours ago, tooth_dr said:

Wow @wimvb that’s incredible! And great data too @HunterHarling  

What a great thread ????

Thanks.

The data certainly is good, but the histogram was very much to the left, and it needed a very hard stretch. There is also some banding which is probably from the camera's read pattern. So, I wonder if, despite the 3 - 5 minutes sub exposures, exposure time could have been longer. Luminance collects the data from three colours at the same time. So, if luminance exposure can be 3 minutes, colour exposure could be 3x3 minutes, with the possible exception of green. Most mono cameras have their highest QE at green. Blue and red are less sensitive. If luminance exposure time is 3 minutes, then green could be at least 6 minutes, and blue and red at least 9 minutes each.

For the ASI1600, with the matching rgb filters, this doesn't apply. For this combination, all three colours should have an exposure time of 9 minutes, if luminance is captured at 3 minutes.

I must test this "theory" coming imaging season. (It's still too bright up here, with only the summer triangle visible at the darkest of night.)

Link to comment
Share on other sites

First off, thanks Rodd for the data. I have to say that I really struggled with this one. Its the first time I have used lrgb and Ha. Just using the RGB I failed to get good colour but after much playing I manged an image that I feel I can show. Still not 100% happy with it but it is what it is.

 

42567661445_59fb2be608_b.jpg

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • 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.