ST1.7 Decon and ST1.8 SV Decon

[Startrek]

Here’s some Crop section images using ST 1.7 Decon and ST1.8 SV Decon as suggested by Ivo

Additional info to follow

Clear Skies
Martin

[Startrek]

More images

Martin

[Startrek]

Final image

Martin

[Startrek]

Images are cropped areas of my recent M8 Lagoon Nebula with various Binning applied to show the results

My skies are Bortle 8 heavy LP so hence I use the new Antlia ALP T OSC dual band filter 5nm

As you can see ST 1.7 Decon works ok and Stars are not greatly affected
ST 1.8 SV Decon performs better at reducing atmospheric blur than ST 1.7 Decon , but Stars are severely affected

In ST 1.8 SV Decon, Binning severely affects Stars
In ST 1.7 Decon, Binning doesn’t affect Stars to much

My image scale is 0.86 arc sec per pixel ( oversampled)

In ST 1.8 SV Decon ( Sample method ) I have tried the following to improve Stars
Carefully select sample stars and use correct sample area
Use up to 10 samples across image
Reduce Sampled iterations to 5x
Use PSF Resampling methods

The above does not improve the “look” of my Stars

So I’m left with either using ST1.8 SV Decon Synthetic mode or go back to ST 1.7 Decon

Comments, Advice, Solutions most welcome

Clear Skies
Martin

[Mike in Rancho]

Interesting, Martin.

Do you have any parameters that were used in either 1.7 or 1.8? Or the sample stack?

I see a couple tiny flaws in star cores, and a wee bit of shape issue (though nothing like I can induce with SVD in my data!), but mostly all I see is lack of deringing.

Now, the question becomes, with the appropriate level of deringing (strength, sensing, and fuzz) to clear that up, then what is the 1.7 and 1.8 comparison as to the detail resolving?

Also, even once you bump up the deringing options, double check the mask - as that is useful for deringing. Stars with saturation problems are sometimes not properly and fully caught (i.e. full stellar profile) by the auto-apod.

Other than that, I'll await Ivo's far better analysis and commentary!

[admin]

Hi Martin,

Would you be able to share the linear bin 50% crop with me/us as well as the processing log that got you the ST 1.8 SV Decon Bin 50% image?
For one, I'm keen to understand what might be causing the severe ringing around HD164816 and the core corruption in some other stars. I haven't really seen this before...

[admin]

Other than that, I'll await Ivo's far better analysis and commentary!

Actually - your commentary is far more useful here!

I *have* seen this before I just realised, and this can indeed happen if stellar profiles of stars that show the ringing are - for whatever reason - not fully included in the mask. First thing to do then is check the mask being generated around the offending area(s).

StarTools not being able to generate a good apodization mask, may be a side-effect of some other issue (for example stacking frames of greatly varying quality, or stacking frames of different exposure lengths), causing star cores to exhibit aberrant data other than overexposure.

[Startrek]

Thanks Mike and Ivo

Below is some more information plus attached photos of masks

Firstly all subs are of equal length ( 2 min )
Sub quality in DSS ranged from 750 to 350 ( discarded anything lower which about 12 frames I think )

I did run same through processing again using Bin 35% , Bin 50% and Bin 71%

With Bin 35% , the Apod Mask looked ok no major stars missing , however Bin 50% and Bin 71% has a poor mask with quite a few stars missing or half masked including the Star in question
See attached photos

M8 Lagoon Workflow parameters for the test samples I sent
Open
Linear from OSC
Ok
AutoDev
Keep
Bin 35% , Bin 50% , Bin 71%
Crop edges
NB Wipe Dark Anomaly Filter 5 pixels
AutoDev ROI , ignore fine detail 4.5 pixels
HDR Tame gamma highlight 1.06
Sharpen DSO amount 150%
SV Decon extra sensitive mask
De noise 6.0 pixels
Crop test area
Save

As far as a log file , I don’t connect my processing laptop to the internet ( only for software upgrades )
I assume the log files have to be uploaded , I’ve never used or looked at one ???


Cheers
Martin

[hixx]

Hi Martin,
this seems to turn into a great discussion, it might help to get to the bottom of this.
The log file (of your last processing should be stored in the Star tools folder and is named "StarTools.log"

Note1 : SVDecon 1.8 synthetic Mode is pretty much identical to Decon 1.7 if no sample stars are used.
Note2 Mike is right, it might be the deringing parameters come into play here. Possibly the 1.8 Deringing default is a bit weaker then 1.7. i use to crank up Deringing Amount to 0.9 and Deringing Detect to 55%-60%. You might want to try out if that improves Your stars.
Clear Skies,
Jochen

[Startrek]

Hi Martin,
this seems to turn into a great discussion, it might help to get to the bottom of this.
The log file (of your last processing should be stored in the Star tools folder and is named "StarTools.log"

Note1 : SVDecon 1.8 synthetic Mode is pretty much identical to Decon 1.7 if no sample stars are used.
Note2 Mike is right, it might be the deringing parameters come into play here. Possibly the 1.8 Deringing default is a bit weaker then 1.7. i use to crank up Deringing Amount to 0.9 and Deringing Detect to 55%-60%. You might want to try out if that improves Your stars.
Clear Skies,
Jochen

Thanks Jochen
Yes great discussion but a frustrated ST 1.8 SV Decon user

I’m certainly not an expert on the computer and had to get my 40 year old son to show me where the log files were located
They were located in my C drive under Programs ( 86x ) , Startools, distro , Startools text files

I will attach the Bin 50% log file tomorrow as requested by Ivo
To be honest I find ST1.7 Decon superior to ST1.8 SV Decon Synthetic option

The more I Bin and improve SNR , the more my Stars turn blocky when I zoom in ( which is perfectly understandable) less pixels per area of Star core
The same goes for SV Decon Sample method , the more I Bin , say from 50% down to 35% , the more the Stars are more blocky due to less pixels covering the specific area of the Star core
My question is why my Apod Mask changes gets worse with 50% and 71% Binning ( I thought it would be the reverse and 35% Bin would not give a good Apod Mask ) ?? Should the Star Mask algorithm compensate for Binning ? Also there’s no parameter to touch up your Apod Mask if Stars are missed or half covered ( this feature is only in a standard Mask where you can use Grow or Shrink etc…. )

Cheers
Martin