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_banner_terminator_challenge_winners.thumb.jpg.6becf44442bc7105be59da91b2bee295.jpg

Rodd

M81-Green channel

Recommended Posts

Here is my next project.  I was shooting subs on this while waiting for M51 to climb higher in the sky.  I am hoping to spot where my data takes a wrong turn.  So far, I think the green channel is done and looks good.This is 34 10 min subs with the C11Edge and STT 8300.  Anybody see anything glaring that will degrade the LRGB, ;let me know. This stack has been lightly processed sop it stands alone.  I was surprised at how big this galaxy is--almost fills the FOV. 

Green-34-2.thumb.jpg.66022c436ad3f017e27d1f2443724564.jpg

  • Like 6

Share this post


Link to post
Share on other sites
10 minutes ago, Daniel-K said:

and still my most favorite galaxy.  that looks really nice well done 

Thanks Danny.  I think I uploaded the wrong one--for a stand alone the background should be a bit darker.

Green-34-2a.thumb.jpg.ad036e22caf6d4fab0e9a0ae76ed1935.jpg

  • Like 2

Share this post


Link to post
Share on other sites

Loads of detail, nice depth to the data.

Are you using the 0.7 reducer too?

When you say "wrong turn", how does that manifest itself?

Share this post


Link to post
Share on other sites
3 hours ago, Tim said:

Loads of detail, nice depth to the data.

Are you using the 0.7 reducer too?

When you say "wrong turn", how does that manifest itself?

Thanks Tim.  Yes I am using the .7x reducer.  I haven't tried the scope for deep space without it, only the Moon and planets. I don't know often times how the "wrong turn" manifests itself.  usually in the processing of the data frankly.  But my M51 is a good example--the subs looked good and the stacks looked good, but I am not satisfied with the end result even though the consensus is it is pretty good.  Ot the Leo Triplet image--29 hours of data and images made with 3-4 hours look better to me.

Rodd

Share this post


Link to post
Share on other sites
3 hours ago, artem said:

Looks really nice :icon_albino:

Thanks artem.

Share this post


Link to post
Share on other sites

That looks great. Hopefully you'll get the other colours done before clouds spoil your efforts.

Judging from the stars in this image and your M51, I think that your biggest enemy is seeing. There's a slight haze around the brighter stars.

Deconvolution in the linear stage can clean this up (somewhat).

Share this post


Link to post
Share on other sites

Looking good Rodd.

One note: watch that you don't inadvertantly modify the core to give it a hard-edged circle/spherical artefact as this wouldn't be reflected in the reality; I think you may avoid this when you add luminance, however, you are getting close to it.

Share this post


Link to post
Share on other sites
6 hours ago, Barry-Wilson said:

Looking good Rodd.

One note: watch that you don't inadvertantly modify the core to give it a hard-edged circle/spherical artefact as this wouldn't be reflected in the reality; I think you may avoid this when you add luminance, however, you are getting close to it.

Yeah I know--I saw that.  I will be more careful when I process the image for real.  I am pleased with the green channel though.  I hope the other channels come out as good. 

Share this post


Link to post
Share on other sites

The funny thing with green is that you apply scnr to it to reduce it. Seems a shame to me, seeing your excellent result.

(He said, running for cover.)

 

Share this post


Link to post
Share on other sites
11 hours ago, wimvb said:

The funny thing with green is that you apply scnr to it to reduce it. Seems a shame to me, seeing your excellent result.

(He said, running for cover.)

 

I know--that does seem strange.  Maybe that's why though--the green comes out bright and with lots of signal, overpowering the others (just a tongue in cheek thought--maybe total hogwash.

You mentioned deconvolution in the linear state--When ever I try to do anything to linear images (other than BN, CC, DBE and stretching) nothing seems to happen.  Is duplicating the image, making teh duplicate non linear so as to make a mask, then using that to mask the linear image so as to protect everything but the stars the way to do it for deconvolution?

I am always wondering if its the seeing--which usually is bad to poor these days, or focus.  

Share this post


Link to post
Share on other sites
4 hours ago, Rodd said:

...

You mentioned deconvolution in the linear state--When ever I try to do anything to linear images (other than BN, CC, DBE and stretching) nothing seems to happen.  Is duplicating the image, making teh duplicate non linear so as to make a mask, then using that to mask the linear image so as to protect everything but the stars the way to do it for deconvolution?

...  

I try to do as much as possible in the linear stage. I got the tip of Alejandro Tombolini to try to fix problems before they become problems. For example, try to keep stars down before they get bloated in stretching. So here's my workflow (adapted to my light polluted imaging location)

registering, and integration in BPP script in PI

dynamic crop to get rid of stacking edges and severe amp glow on the left hand side of my images

DBE, depending on the image, either with few or many samples. I try to use the largest sample size (pixels) that will fit between stars.

Background Neutralization

HSV repair script. DBE takes out a lot of the sky glow and light pollution. But this gives false colours in the brightest stars. HSV repair script will correct this. I do this before determining the white point, because it will affect the results of colour calibration.

Colour calibration

Deconvolution, mainly to make stars narrower. Deconvolution will restore star shape to the standard bell curve shape. You won't see this in the screen stretched image, but if you remove the STF, you'll see significantly narrower stars. Small stars are a little more peaked as well. Depending on the regularization layers and settings, this can also lift detail in galaxies and nebulae. I've never had much luck with deconvolution until I started guiding. Now I have much better star shapes, which respond well to the process.

Noise reduction (TGVDenoise is my weapon of choice). Previously I only used it on L in L*a*b mode. But with the new TGV process, I start to use it on chrominance as well. I usually tame the process by using a 50% mask. TGV noise reduction exels at removing small scale noise. To remove large scale colour mottle (chroma noise), I use MMT on 7 layers, in Chrominance (Restore Y) mode.

Histogram transformation or masked stretch. Masked stretch results in smaller stars, but also more noise.

This ends the linear process.

The nonlinear process is mainly about stretching, colour saturation, etc. Occasionaly HDR transform and LocalHistogramEqualization

When the image is almost complete, I create a contour star mask and use Morphology Transform to pinch the stars, making them smaller. Then I apply MLT with bias in the 2nd, 3rd, and 4th layers (with a starmask in place) This process will create small bright stars.

Writing all this, I realise I must do a proper write up sometime.

 

Hope this helps.

  • Like 1

Share this post


Link to post
Share on other sites
25 minutes ago, wimvb said:

I try to do as much as possible in the linear stage. I got the tip of Alejandro Tombolini to try to fix problems before they become problems. For example, try to keep stars down before they get bloated in stretching. So here's my workflow (adapted to my light polluted imaging location)

registering, and integration in BPP script in PI

dynamic crop to get rid of stacking edges and severe amp glow on the left hand side of my images

DBE, depending on the image, either with few or many samples. I try to use the largest sample size (pixels) that will fit between stars.

Background Neutralization

HSV repair script. DBE takes out a lot of the sky glow and light pollution. But this gives false colours in the brightest stars. HSV repair script will correct this. I do this before determining the white point, because it will affect the results of colour calibration.

Colour calibration

Deconvolution, mainly to make stars narrower. Deconvolution will restore star shape to the standard bell curve shape. You won't see this in the screen stretched image, but if you remove the STF, you'll see significantly narrower stars. Small stars are a little more peaked as well. Depending on the regularization layers and settings, this can also lift detail in galaxies and nebulae. I've never had much luck with deconvolution until I started guiding. Now I have much better star shapes, which respond well to the process.

Noise reduction (TGVDenoise is my weapon of choice). Previously I only used it on L in L*a*b mode. But with the new TGV process, I start to use it on chrominance as well. I usually tame the process by using a 50% mask. TGV noise reduction exels at removing small scale noise. To remove large scale colour mottle (chroma noise), I use MMT on 7 layers, in Chrominance (Restore Y) mode.

Histogram transformation or masked stretch. Masked stretch results in smaller stars, but also more noise.

This ends the linear process.

The nonlinear process is mainly about stretching, colour saturation, etc. Occasionaly HDR transform and LocalHistogramEqualization

When the image is almost complete, I create a contour star mask and use Morphology Transform to pinch the stars, making them smaller. Then I apply MLT with bias in the 2nd, 3rd, and 4th layers (with a starmask in place) This process will create small bright stars.

Writing all this, I realise I must do a proper write up sometime.

 

Hope this helps.

Thanks a bunch Wim--Most of it is part of my flow too--except the HSV script--I will have a go at that, and will try decon on stars in linear state.  Not sure how you can see the result though if you remove the STF--if you have to wait until you stretch the image it seems problematic. 

Share this post


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

Thanks a bunch Wim--Most of it is part of my flow too--except the HSV script--I will have a go at that, and will try decon on stars in linear state.  Not sure how you can see the result though if you remove the STF--if you have to wait until you stretch the image it seems problematic. 

Even without screen stretch, you can see the brightest stars (bloated) and the somewhat dimmer stars. You can also use the wrench tool in STF to create your own screen stretch, and apply a milder screen stretch.

Share this post


Link to post
Share on other sites

Here are unstretched versions of an image showing the effect of HSV repair, and the effect of deconvolution on stars.

First the uncorrected version of the image. Left: linear; right: stretched. Note that the stars have near identical cores, but different halos and diffraction spikes

StarColour_uncorrected.thumb.jpg.21b99662290ff390cbf27732c5dab3ad.jpg

HSV repair: Left is the unrepaired original, right is the repaired version. Again, look at the star colour of the larger stars. Images are not stretched

Uncorr_HSVCorr.thumb.jpg.7ad434b53f1cf617bcc83884245fe952.jpg

Next deconvolution. The effect is subtle, but makes all the difference later on in processing. The larger stars are just a little narrower. The smaller stars are sharper and slightly brighter. The smallest stars are better defined.

Before_AfterDeconvolution.thumb.jpg.017bb28a8abf0e738e7f2cea2183d49f.jpg

 

  • Like 1

Share this post


Link to post
Share on other sites
2 hours ago, wimvb said:

Here are unstretched versions of an image showing the effect of HSV repair, and the effect of deconvolution on stars.

First the uncorrected version of the image. Left: linear; right: stretched. Note that the stars have near identical cores, but different halos and diffraction spikes

StarColour_uncorrected.thumb.jpg.21b99662290ff390cbf27732c5dab3ad.jpg

HSV repair: Left is the unrepaired original, right is the repaired version. Again, look at the star colour of the larger stars. Images are not stretched

Uncorr_HSVCorr.thumb.jpg.7ad434b53f1cf617bcc83884245fe952.jpg

Next deconvolution. The effect is subtle, but makes all the difference later on in processing. The larger stars are just a little narrower. The smaller stars are sharper and slightly brighter. The smallest stars are better defined.

Before_AfterDeconvolution.thumb.jpg.017bb28a8abf0e738e7f2cea2183d49f.jpg

 

Thanks Wim--I have used Decon to tighten up stars but in the non linear state.  It works, but I don't do it all that often.  I do use Decon with shape and deviation set to 0.6 as a sharpening tool--works quite well to get a little sharpening to fine details.  Sometimes it creates artiifacts though--it works about half the time.  I will have to try and incorporate this into my linear work flow.  I took a look at the HSV script--there are several settings and chjoices that I do not get yet.  I will try and incorporate that as well.

Share this post


Link to post
Share on other sites

Just use std settings, but check unrepaired V box. Use this later in combination of channels.

Share this post


Link to post
Share on other sites
57 minutes ago, sulaco said:

Incredible image in even just one channel

Thanks sulaco.  Can't wait to get teh otehrs.

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.