Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Processing skills ?

  • Please log in to reply
68 replies to this topic

#1 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 08:57 AM

Been messing around with the hobby for a few years though never really developed any processing skills, only DSS and LR. Here is an image of the Iris, 82x180s with a stock DSLR with 10 darks and master bias. I believe there is a lot more data in the image that can be processed with better software and skill. Would like to see someone have a go at it.  https://drive.google...iew?usp=sharing

 

ris22 77.jpg


  • kathyastro likes this

#2 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 09:58 AM

Been messing around with the hobby for a few years though never really developed any processing skills, only DSS and LR. Here is an image of the Iris, 82x180s with a stock DSLR with 10 darks and master bias. I believe there is a lot more data in the image that can be processed with better software and skill. Would like to see someone have a go at it.  https://drive.google...iew?usp=sharing

 

attachicon.gifris22 77.jpg

Edit:  Autosave seems to be working fine when saved as image.  Working on it now in Startools.


Edited by GoldSpider, 28 February 2020 - 10:21 AM.


#3 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,914
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 28 February 2020 - 10:07 AM

I could be wrong, but the Autosave that DSS produces is stretched and not linear.

It depends on whether DSS was being used to stack linear files (e.g. raw files) or non-linear files (e.g. TIFs created by Lightroom).

 

I explained it all in my post yesterday:

 https://www.cloudyni...the-saved-file/

 

Maybe the OP could confirm.

 

The autosave file does appear to be rotated by about 45 degrees though, compared with the image in the original post, which is odd. 

It also has a lot of vignetting because no flats were used.

 

Mark


Edited by sharkmelley, 28 February 2020 - 10:30 AM.


#4 Gipht

Gipht

    Vanguard

  • *****
  • Posts: 2,201
  • Joined: 12 Nov 2016
  • Loc: Prescott Valley, AZ.

Posted 28 February 2020 - 10:11 AM

Here is a quick run through taking the autosave into DSS and then saving as a picture file.  This is from Star Tools.  Very similar to your results.

 

 

Attached Thumbnails

  • Capture.JPG

  • zxx and kathyastro like this

#5 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 10:39 AM

You lose some data to cropping when some frames are shot at a different orientation, but what you had that didn't need cropped was quite nice.  Here's what I was able to do with it using Startools. 
 
Read noise (not an uncommon issue with an uncooled camera) limits how much detail can be wrung out of an image.  What ISO were you shooting at?  Were you dithering?
 
iris Zxx

  • zxx and Swordfishy like this

#6 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 11:11 AM

It depends on whether DSS was being used to stack linear files (e.g. raw files) or non-linear files (e.g. TIFs created by Lightroom).

 

I explained it all in my post yesterday:

 https://www.cloudyni...the-saved-file/

 

Maybe the OP could confirm.

 

The autosave file does appear to be rotated by about 45 degrees though, compared with the image in the original post, which is odd. 

It also has a lot of vignetting because no flats were used.

 

Mark

They are raw files from Pentax K50 stacked in DSS, maybe had settings wrong. Correct, no flats


Edited by zxx, 28 February 2020 - 11:20 AM.


#7 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 11:17 AM

 

You lose some data to cropping when some frames are shot at a different orientation, but what you had that didn't need cropped was quite nice.  Here's what I was able to do with it using Startools. 
 
Read noise (not an uncommon issue with an uncooled camera) limits how much detail can be wrung out of an image.  What ISO were you shooting at?  Were you dithering?

 

 

ISO 800, no dithering because my K50 cannot be tethered to the PC with capture software. Guess I could manually dither.



#8 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,914
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 28 February 2020 - 11:26 AM

They are raw files from Pentax K50 stacked in DSS, maybe had settings wrong. Correct, no flats

So the autosave.tif file is linear. So it looks just fine when opened by Photoshop in 32-bit mode.  The background subtraction is awkward though because of the vignetting.

 

Mark


  • zxx likes this

#9 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 11:28 AM

ISO 800, no dithering because my K50 cannot be tethered to the PC with capture software. Guess I could manually dither.

 

Same limitation I have in shooting with my Sony mirrorless.  There's probably a way to set manual dithering in PHD2 on a timer that you can match up to your intervalometer.


  • zxx likes this

#10 ChristopherBeere

ChristopherBeere

    Mariner 2

  • -----
  • Posts: 214
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 28 February 2020 - 11:28 AM

This is not a properly calibrated data set and the signal to noise ratio is mediocre.

 

Respectfully, no amount of processing tricks will make this look any better than your original.


Edited by ChristopherBeere, 28 February 2020 - 11:31 AM.


#11 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 11:29 AM

So my guess is the data is not so good ?  Processing software and skill won't really help.



#12 ChristopherBeere

ChristopherBeere

    Mariner 2

  • -----
  • Posts: 214
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 28 February 2020 - 11:39 AM

So my guess is the data is not so good ?  Processing software and skill won't really help.

 

The data is reasonable but if you want to expose the faint peripheral dust you need a lot more integration time to enable you to get it above the noise ceiling with aggressive stretching.

 

Your processing effort is actually pretty impressive given the quality of the data.


  • zxx, GoldSpider and kathyastro like this

#13 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 11:58 AM

Thanks all for the time messing around with the data. Guess if I really want to improve it's time to move up to a OSC cooled camera ? Or maybe a modified D5300.


Edited by zxx, 28 February 2020 - 12:02 PM.


#14 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 12:01 PM

The data is reasonable but if you want to expose the faint peripheral dust you need a lot more integration time to enable you to get it above the noise ceiling with aggressive stretching.

 

Your processing effort is actually pretty impressive given the quality of the data.

Agreed!



#15 ChristopherBeere

ChristopherBeere

    Mariner 2

  • -----
  • Posts: 214
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 28 February 2020 - 12:20 PM

Thanks all for the time messing around with the data. Guess if I really want to improve it's time to move up to a OSC cooled camera ? Or maybe a modified D5300.

 

A modded camera will make a big difference with H-Alpha emission lines at ~656nm  but in this specific instance the Iris is a reflection nebula and has low emission in those spectral lines so it wouldnt actually help that much.

 

Definitely worth getting it modded though as the majority of deep sky objects have strong H-Alpha emission.You will see a big difference.

 

To improve your data you need skies with a super low background sky brightness level (MPSAS > 22.00) low humidity and excellent transparency. 

 

Hardware and software will only get you so far. Excellent data is fundamentally dictated by the conditions you capture under. 

 

Thats not to say you cant achieve great results under average skies but photometric quality skies yield mint data and make processing a simple case of putting the cherry on top. There are always challenges even with top data but life is made a lot easier when you have mega SNR.


Edited by ChristopherBeere, 28 February 2020 - 12:35 PM.

  • zxx likes this

#16 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 12:35 PM

A modded camera will make a big difference with H-Alpha emission lines at ~656nm  but in this specific instance the Iris is a reflection nebula and has low emission in those spectral lines so it wouldnt actually help that much.

 

Definitely worth getting it modded though as the majority of deep sky objects have strong H-Alpha emission.You will see a big difference.

 

To improve your data you need skies with a super low background sky brightness level (MPSAS > 22.00) low humidity and excellent transparency

 

Hardware and software will only get you so far. Excellent data is fundamentally dictated by the location and conditions you capture under and clinical calibration.

 

Thats not to say you cant achieve great results under average skies but photometric quality skies yield mint data and make processing a simple case of putting the cherry on top.

My problem with the K50 is that I can't tether it to the PC, therefor makes dithering difficult. I shot this at a friends house, bortle 4 sky. I should have went 4 or 5 min exposures. most all my imaging has been done from my bortle 8 home sky.  I'm thinking a OSC cooled cam would be the way to go.



#17 ChristopherBeere

ChristopherBeere

    Mariner 2

  • -----
  • Posts: 214
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 28 February 2020 - 12:45 PM

My problem with the K50 is that I can't tether it to the PC, therefor makes dithering difficult. I shot this at a friends house, bortle 4 sky. I should have went 4 or 5 min exposures. most all my imaging has been done from my bortle 8 home sky.  I'm thinking a OSC cooled cam would be the way to go.

 

A cooled camera will help with consistent calibration and noise but you are still at the mercy of the sky background brightness level.

 

Narrowband imaging with a mono camera would yield much better results under suburban skies.



#18 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 12:55 PM

A cooled camera will help with consistent calibration and noise but you are still at the mercy of the sky background brightness level.

 

Narrowband imaging with a mono camera would yield much better results under suburban skies.

Yes I thought about that, though I will be moving soon to a dark country sky. I like to keep it as cheap and simple as possible.



#19 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 12:59 PM

Yes I thought about that, though I will be moving soon to a dark country sky. I like to keep it as cheap and simple as possible.

I've been working with a cooled mono camera for a few months, and I've come to appreciate "simplicity" more since I sold the ASI294MC-Pro.  When everything works well, mono produces a beautiful image, but there are many more points of failure.  I'm most likely going to be switching back to OSC.  Your mileage may vary, of course.


  • zxx and ChristopherBeere like this

#20 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,914
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 28 February 2020 - 01:02 PM

Thanks all for the time messing around with the data. Guess if I really want to improve it's time to move up to a OSC cooled camera ? Or maybe a modified D5300.

The data quality is not too bad.   A modified camera will make little difference to this object because it is a reflection nebula.  The biggest difference you can make is to use flat frames for calibration - this will help enormously.

 

Mark


Edited by sharkmelley, 28 February 2020 - 01:03 PM.

  • Joe G, zxx and ChristopherBeere like this

#21 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 01:12 PM

The data quality is not too bad.   A modified camera will make little difference to this object because it is a reflection nebula.  The biggest difference you can make is to use flat frames for calibration - this will help enormously.

 

Mark

That's my laziness, guess I could still make some flats. I did clean the sensor before this image so dust should not be an issue.



#22 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,914
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 28 February 2020 - 01:18 PM

That's my laziness, guess I could still make some flats. I did clean the sensor before this image so dust should not be an issue.

Give it a try - you have nothing to lose.

 

Mark


  • zxx likes this

#23 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 01:20 PM

That's my laziness, guess I could still make some flats. I did clean the sensor before this image so dust should not be an issue.

 

The stretched data definitely shows some dust on the sensor.  Again, something easily corrected with flats.  It's worth the time every time.

 

The stretched stack also shows that the light frames were shot at two different angles.  Did you rotate the camera at some point during your image capture, or shoot over two nights?  


  • ChristopherBeere likes this

#24 zxx

zxx

    Apollo

  • -----
  • topic starter
  • Posts: 1,360
  • Joined: 27 Jan 2010

Posted 28 February 2020 - 01:29 PM

The stretched data definitely shows some dust on the sensor.  Again, something easily corrected with flats.  It's worth the time every time.

 

The stretched stack also shows that the light frames were shot at two different angles.  Did you rotate the camera at some point during your image capture, or shoot over two nights?  

Two nights, yes should have taking flats. Knew I was gonna crop it down. My 0.8x FR is not a good match for my SW ED80 so I always crop it out. Should of brought My 6RC.  I also found that spot remover in LR does a good job with dust spots   lol


Edited by zxx, 28 February 2020 - 01:44 PM.


#25 GoldSpider

GoldSpider

    Viking 1

  • *****
  • Posts: 928
  • Joined: 20 Apr 2015
  • Loc: Pennsylvania

Posted 28 February 2020 - 01:48 PM

Two nights, yes should have taking flats. Knew I was gonna crop it down. My 0.8x FR is not a good match for my SW ED80 so I always crop it out. Should of brought My 6RC.  I also found that spot remover in LR does a good job with dust spots   lol

 

There's no problem with taking shots over multiple nights, just try to get the orientation close.  That's less of the frame that needs to be cropped out.


  • zxx likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics