Jump to content

Stargazers Lounge Uses Cookies

Like most websites, SGL uses cookies in order to deliver a secure, personalised service, to provide social media functions and to analyse our traffic. Continued use of SGL indicates your acceptance of our cookie policy.

sgl_imaging_challenge_celestial_motion.thumb.jpg.a9e9349c45f96ed7928eb32f1baf76ed.jpg

FaB-Bo-Peep

M51, what can you do with this stacked Tiff please?

Recommended Posts

Oh and you don't stop learning, revisiting previous data with new knowledge can pull out more, I found that on comet data from 2 years ago revisited.

Share this post


Link to post
Share on other sites
Posted (edited)

The latest bleeding edge version I generally find very stable so always run what ever that is and I thought Ivo's very recent YouTube video a good start too as well as the blog mentioned above.

Calibration should help create a better background, I found it was either noisy or close to clipped on processing it.

I forgot to specifically work on drawing the colour out, oh well.

1831848852_Autosave(3.1).png.9b94d7d3f2918329d3998ee5eaf2cf8e.png

 

Edited by happy-kat
  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)
On 15/05/2019 at 20:12, wimvb said:

 

Process:

  • crop all stacking artefacts
  • DBE in several stages. First with a few markers to remove the LP gradient. Then with more markers to remove the vignetting. Finally a pass to reduce some of the remaining artefacts.
  • Photometric Colour Calibration, including background neutralisation
  • Extraction of Luminance
  • Arcsinh and masked stretch of the colour image.
  • Various passes of colour saturation with mask to protect the background

On the synthetic luminance:

  • deconvolution (this increased the noise in the lighter parts of the galaxy, but I wanted to reveal more structure here)
  • stretching
  • HDR compression
  • noise reduction (with a 50% lightness mask)

Then LRGB combination followed by chrominance noise reduction.

  • Curves transformation, targeting the b-component in Lab mode, with a lightness mask to protect the background. This step enhances the blue colour in the galaxy
  • Chrominance noise reduction of the background

 

7

Wim  (wimvb)  thanks for sharing your processing steps.  The end result you achieved is simply amazing.

Can I ask a couple of questions on what you did?

 

How did you extract Luminance, simply clone the image and convert to grayscale or extract Lightness or some other method?

What process did you use to do the noise reduction on the Luminance ?

 

I am not in your league using PI but keen to learn.

Edited by wornish

Share this post


Link to post
Share on other sites
1 hour ago, wornish said:

Can I ask a couple of questions on what you did?

 

How did you extract Luminance, simply clone the image and convert to grayscale or extract Lightness or some other method?

What process did you use to do the noise reduction on the Luminance ?

In pixinsight  there's a process button that will extract luminance. You can also use the process channel extraction in either Lab mode or Lch mode, with only L selected. Before I did this I equalised the channels. Set the rgb sliders in colour workspace to 1 (process menu - colorspace)

Noise reduction: I tried the sigma thresholding option in multiscale linear transformation, but I may have ended up using the standard noise reduction in MLT with an inverted lightness mask. I would have to check the process history to make sure which I used. I actually used this image to try a few new tricks. My standard noise reduction on L is either TGV denoise or MLT, applied with an inverted L mask. I always increase the background of the L mask to 50%. This way I get noise reduction, and not noise obliteration. Hope this helps.

  • Thanks 1

Share this post


Link to post
Share on other sites
15 hours ago, FaB-Bo-Peep said:

 My stack comprises of Large JPEGS so I will need to make sure my DSLR is switched to RAW files for my next project and look at taking some flats. Just had another go this time using a 32 bit FITS file saved from DSS as apposed to a 16 bit TIFF, the file is much larger so must contain more data to start with yes?

Aha! That explains a lot :) JPEG is an 8-bit lossy image file format, where you camera applies all sorts of processing (stretching, color calibation, sharpening, sometimes noise reduction) before saving it to the card. You'll want to avoid all this. When processing this dataset, however, you can make use of StarTools' ability to "reverse" the stretch JPEG encoding applied (choose "Non-linear sRGB source" when opening the dataset), so at least you can work with somewhat linear data. StarTools' modules should then do a better job across the board.

Honestly though, when you manage to grab signal like this in mere lossy 8-bit JPEGs, this bodes very well for your future endeavours!

Saving as FITS is the favourite format when it comes to astrophotography datasets, as it is "philosophically" much closer to scientific instrument data, and virtually never used for finished images meant for human consumption. Because you stacked 8-bit JPEGs you probably won't see any fidelity gains using 32-bit FITS, but using a 32-bit FITS file format (Integer is best) for your datasets is good practice.

Quote

To me it looks like another step in the right direction but it's all subjective and as each time I process in Startools I'm using slightly different settings, tweaks and the masks so obviously it's impossible to exactly replicate previous results, someone could go mad with all this tweaking, does anyone ever truly finish an image?

You can find all the steps you took in a file called StarTools.log (should be in the same folder as the executables), so you can precisely replicate them if you need to. As of the 1.4.x beta versions, it also stores the masks in that file, so you can recover them too. They are stored in BASE64 format and look like long text strings. For more info on how to convert these strings back to PNGs you can import as masks, have a look on the StarTools website. (I'm the author of StarTools, therefore I don't think it's appropriate to "spam" direct links to my own website, but it should be easy to find)

Getting "closure" on an image becomes easier once you get to know you gear and the characteristics of the datasets you produce with it. You'll start getting a sense of what you "can get away with", how far you can push things, but also - and that's what this hobby is all about - your personal tastes. Also not unimportant - to some - is being able to document and replicate faithfully the physical & chemical properties of the objects you are imaging. Emissions, reflections, temperatures, shockwaves, tidal tails. All objects have a story; a past, a present, a future.

  • Thanks 1

Share this post


Link to post
Share on other sites
5 hours ago, jager945 said:

I don't think it's appropriate to "spam" direct links to my own website, but it should be easy to find)

Give yourself a plug. Your app is going places:)

https://www.startools.org/

  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)
17 hours ago, jager945 said:

you can make use of StarTools' ability to "reverse" the stretch JPEG encoding applied (choose "Non-linear sRGB source" when opening the dataset), so at least you can work with somewhat linear data. StarTools' modules should then do a better job across the board.

So much useful info coming in as a result of this post and I'm so very grateful to all that have taken the time to reply. I gave the above a quick go and below is the result. It was only a quick go but again the result is very pleasing to me. I must say I'm finding Startools an amazing program for a novice like me, I'm sure there's a lot of complex stuff going on in the background but the user interface makes processing images so easy without needing a degree in Photoshop. I'm sure I'm learning more with each attempt but compared to my last go a few years back it's got to the stage where I am pleased with all the versions I'm ending up with now. Can't wait to see what my next target comes out like using RAW files 🙂 

m51-fits-32-integer-rgbimport-jpeg.jpeg

Edited by FaB-Bo-Peep
  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)

And a quick play with curves in Paintshop Pro to darken the Sky a bit.

 

m51-fits-32-integer-rgbimport-jpeg-jpg.jpg

Edited by FaB-Bo-Peep

Share this post


Link to post
Share on other sites

If you roll the colour dropper toll over the sky background you'll see if you have now clipped it, when the background is reading 0.

Share this post


Link to post
Share on other sites

Not a DSO astro imager but I can appreciate the help everyone has given to improve a good image into a great image. Well done all. Des

Share this post


Link to post
Share on other sites
On ‎18‎/‎05‎/‎2019 at 02:01, jager945 said:

Aha! That explains a lot :) JPEG is an 8-bit lossy image file format, where you camera applies all sorts of processing (stretching, color calibation, sharpening, sometimes noise reduction) before saving it to the card. You'll want to avoid all this.

Thanks so much for the heads up on this. Last night I had a go at capturing M13, this time with my camera set to RAW and although I only had time for a quick play with the data this morning before leaving for work the difference in detail and colour is amazing!  I will have a proper play tonight and post the end result in the imaging DSO section 🙂

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.