Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Astro Pixel Processor vs DSS thoughts

PeterCPC

By PeterCPC
in Discussions - Software

Recommended Posts

  • Replies 40
  • Created
  • Last Reply
gnomus

gnomus

Posted
Posted
16 minutes ago, PeterCPC said:

After getting lots of info from Mabula re settings etc in APP, I did another run through with the same (not very good) data. 23 lights, 25 Bias, 26 Darks and 8 Flats.

This was the final stack from APP (no post processing) and you can see some strange artifacts (wiggly lines) and you can still see some satellite trails. The DSS result did not show any of these. I have asked Mabula about these but no reply as yet.

I do like the Bad Pixel Map that APP produces and the easy Background neutralisation. DSS sometimes throws a wobbly so it might be worth having APP as well.

I downloaded the trial of AA6 but am struggling with it without any manual - I use Canon CRAW images but the colour seems all wrong.

Peter

St-avg-6900.0s-NR-x_1.0_LZ3-NS-full-eq-add-sc_BWMV_nor-AA-RL-MBB5_1stLNC_it1-St.jpg

Hi 

The image looks pretty good.  My 'guesses' as to the artefacts are as follows.  The ring of dots to the right of the galaxy look like bad pixels, dithered.  You could try redoing the BPM with different settings to see if you could generate a BPM that captures these.  Or you could clone them out, or ...

The squiggly lines and the satellite trail may be due to your integration rejection settings.  Have you compared average sigma vs winsorized sigma?  And have you tried different settings for these algorithms?  More subs would, of course, increase the likelihood of these artefacts being rejected.  .... and better rejection might get rid of your remaining dots (bad pixels) too.

Link to comment
Share on other sites
PeterCPC

PeterCPC

Posted
  • Author
Posted
31 minutes ago, gazza said:

Did the trial you downloaded have the help file? It contains all you need to know...

Gary

Yes the help file is there but I cannot find anything to do with Canon RAW settings - I am pretty sure that the Bayer pattern is RGGB but, when I select that in Options, the colour comes out wrong. I also found that the saturation was way too much.

I don't find the Help file very helpful :icon_biggrin:. I can't find a manual online.

Peter

Link to comment
Share on other sites
PeterCPC

PeterCPC

Posted
  • Author
Posted
20 minutes ago, gnomus said:

Hi 

The image looks pretty good.  My 'guesses' as to the artefacts are as follows.  The ring of dots to the right of the galaxy look like bad pixels, dithered.  You could try redoing the BPM with different settings to see if you could generate a BPM that captures these.  Or you could clone them out, or ...

The squiggly lines and the satellite trail may be due to your integration rejection settings.  Have you compared average sigma vs winsorized sigma?  And have you tried different settings for these algorithms?  More subs would, of course, increase the likelihood of these artefacts being rejected.  .... and better rejection might get rid of your remaining dots (bad pixels) too.

It still is not better than I achieved with DSS. Yes, I'm sure that you are correct about the settings - it's the usual problem of not knowing what settings to use. I'm hoping Mabula will make some suggestions as well.

Peter

Link to comment
Share on other sites
gnomus

gnomus

Posted
Posted
42 minutes ago, PeterCPC said:

It's not being developed anymore .... 

Does that matter, if it is giving you the results you want?  I am on an older version of Photoshop and don't intend to upgrade, because I don't like the new licensing model.

43 minutes ago, PeterCPC said:

.... Thanks for your input.

No worries.

Link to comment
Share on other sites
gnomus

gnomus

Posted
Posted
8 hours ago, PeterCPC said:

It still is not better than I achieved with DSS. Yes, I'm sure that you are correct about the settings - it's the usual problem of not knowing what settings to use. I'm hoping Mabula will make some suggestions as well.

Peter

Incidentally, I was working on some M42 subs today that had bad satellite trails.  On one stack, Sigma rejection didn't work.  I went to Winsorized Sigma and they vanished.  So you could try that.

Link to comment
Share on other sites
gazza

gazza

Posted
Posted
18 hours ago, PeterCPC said:

Yes the help file is there but I cannot find anything to do with Canon RAW settings - I am pretty sure that the Bayer pattern is RGGB but, when I select that in Options, the colour comes out wrong. I also found that the saturation was way too much.

I don't find the Help file very helpful :icon_biggrin:. I can't find a manual online.

Peter

So in the RGB demosaic  dialogue, did you try reducing the saturation - recommended setting is 20-40, Did you try adjusting the white balance setting in the same dialogue?There is no plug and play setting for these, they will all depend on both the target and the equipment used. Learning to process is arduous - it involves a lot of trying stuff out, and finding what doesn't work...None of the advanced programs will produce fabulous output without tweaking settings. Another great program to try is CCDStack (trial available), but this also requires ages to learn, either that or the purchase of tutorials. have a look at a couple of images of a friend of mine who uses cCDStack.....

http://www.slate.com/blogs/bad_astronomy/2015/05/12/astrophoto_terry_robinson_s_carina_nebula_from_melbourne.html ...... he paid for the online tutorials........ :-) There is no way he got anything like that result from DSS

cheers

Gary

Link to comment
Share on other sites
PeterCPC

PeterCPC

Posted
  • Author
Posted
23 hours ago, gnomus said:

Hi 

The image looks pretty good.  My 'guesses' as to the artefacts are as follows.  The ring of dots to the right of the galaxy look like bad pixels, dithered.  You could try redoing the BPM with different settings to see if you could generate a BPM that captures these.  Or you could clone them out, or ...

The squiggly lines and the satellite trail may be due to your integration rejection settings.  Have you compared average sigma vs winsorized sigma?  And have you tried different settings for these algorithms?  More subs would, of course, increase the likelihood of these artefacts being rejected.  .... and better rejection might get rid of your remaining dots (bad pixels) too.

Mabula suggested using the Sigma clip outlier rejection filter with 1 iteration and Kappa set to 2.5. This removed both the satellite trails and the wiggly lines.

Peter

Link to comment
Share on other sites
gnomus

gnomus

Posted
Posted
1 minute ago, PeterCPC said:

Mabula suggested using the Sigma clip outlier rejection filter with 1 iteration and Kappa set to 2.5. This removed both the satellite trails and the wiggly lines.

Peter

Result.  So what were you doing before?

Link to comment
Share on other sites
gnomus

gnomus

Posted
Posted
55 minutes ago, PeterCPC said:

I had not selected the outlier rejection filter at all - you live and learn.

Peter

:icon_biggrin:

I assume that DSS was choosing one form of rejection, by default.  Have you tried the LP, Background Calibration and Star Calibration tools yet?  It might be useful, at some stage, to have an updated image to assist those sitting on the fence about APP.

Link to comment
Share on other sites
PeterCPC

PeterCPC

Posted
  • Author
Posted

Latest image after applying LP, Background calibration and Star calibration. I would do further processing in other packages like Paintshop Pro but this gives an idea of what APP can produce on it's own. Bear in mind that I am still learning APP.

Peter

St-med-6900.0s-SC_1_2.5-x_1.0_LZ3-NS-full-eq-add-sc_BWMV_nor-AA-RL-MBB5_1stLNC_it1-lpc-cbg-csc-St-1.jpg

Link to comment
Share on other sites

Archived

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

Go to topic listing

  • 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.