#1  
Old 09-06-2022, 04:05 PM
Ryderscope's Avatar
Ryderscope (Rodney)
Registered User

Ryderscope is offline
 
Join Date: May 2006
Location: Glanmire, NSW
Posts: 2,168
How much to Zoom?

One thing I have been pondering for a while is to consider how far one should go in zooming into images (a.k.a. pixel peeping), specifically when evaluating a new image (published by others) that we have been presented with. To help with my ponderings I would like to kick off a discussion here to look at this issue from a number of angles and see where it takes us. I would like to 'stir the pot' so to speak and see what different opinions there may be out there (please by kind though ).

How far should we go when pixel peeping (if at all). What zoom level: 0%; 100%; 200%; 300%; more? Why should we do this? I'm guilty of this in that I'm always tempted to zoom to 100% when looking at a new image that others have published with a view to seeing what nasties might lurk therein. But again - why

Things to consider are:

* What is the resolution of the image (this would be based on the camera sensor in terms of pixel size and the physical size of the sensor)?;

* What type of display device is being used to present the image. Possibilities are: 2k 21" monitor; 4k 27" monitor; tablet screen; smartphone; projector; printed format; others?

* Is the image a very high resolution mosaic such that additional detail can be revealed if we zoom into the image?

* The image scale of the system used to generate the image.

* How much resolution can the average human eye see on the screen anyway?

To kick off discussions I will start with a position (which I may, or may not, believe myself ) and see where we go. That position would be to say that most astro images are published with the intent that they are meant to be viewed at the native resolution that they were published in, not at a zoom level of 100% or higher. This being the case it should be redundant to zoom in to the image and find artefacts that cannot be seen at a zoom level where the image fits within the physical confines of the display monitor. In other words, if I cannot see any nasties at normal viewing levels then why should I go looking for them?

What thoughts do others have on this?

Clear skies,
Rodney
Reply With Quote
  #2  
Old 09-06-2022, 04:23 PM
mura_gadi's Avatar
mura_gadi (Steve)
SpeakingB4Thinking

mura_gadi is offline
 
Join Date: Apr 2019
Location: Canberra
Posts: 829
as a non-ap'r

My zooming is limited to getting extra details not as a critiquing exercise, the image should be judge as a whole imo. As a non-ap' person all images deserve a degree of credit...

As for my personnel taste, I must say seeing the big fat Red/Blue and golden stars are the thing! A lot of imagery focuses on nebulosity etc and fail to capture the various star colours around the framing.

Last edited by mura_gadi; 09-06-2022 at 04:36 PM.
Reply With Quote
  #3  
Old 14-06-2022, 11:38 AM
Ryderscope's Avatar
Ryderscope (Rodney)
Registered User

Ryderscope is offline
 
Join Date: May 2006
Location: Glanmire, NSW
Posts: 2,168
Quote:
Originally Posted by mura_gadi View Post
My zooming is limited to getting extra details not as a critiquing exercise, the image should be judge as a whole imo. As a non-ap' person all images deserve a degree of credit...

As for my personnel taste, I must say seeing the big fat Red/Blue and golden stars are the thing! A lot of imagery focuses on nebulosity etc and fail to capture the various star colours around the framing.
Thanks for the response Steve. There are different ways of looking at this when it comes to astro imaging. If one is looking at an image 'as is' on the screen with a view to appreciating it as the author intended, then I could argue that there is no value in zooming in to 300% and then finding artefacts that are not visible otherwise. However if the image resolution supports zooming in for extra details, that is a bonus as we can get extra information that we would not get otherwise. Conversely, if the objective of zooming in is to conduct a technical evaluation by looking for the amount of noise present, chromatic aberrations or other issues, then fine - go for it. Once again though, if one cannot see the issue when displayed to fit within the confines of the display device, why bother?

CS,
Rodney
Reply With Quote
  #4  
Old 14-06-2022, 04:52 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
I'd say 1:1 is as far as you'd go. Good data still holds at that level. I found that if the seeing is good and you dithered doing a drizzle integration will give you a final with details that are still sharp slightly past your image scale so close to 2:1. A good image will look ok at 1:1 and any res lower. Then there's the gear used as well. So although an image might have still edge aberrations if on-axis is excellent then I still see it as an excellent capture because the imperfections are limitations of the gear, not processing induced. Noise, reflections or chromatic aberrations don't bother me the least.
Reply With Quote
  #5  
Old 15-06-2022, 04:09 PM
Ryderscope's Avatar
Ryderscope (Rodney)
Registered User

Ryderscope is offline
 
Join Date: May 2006
Location: Glanmire, NSW
Posts: 2,168
Quote:
Originally Posted by multiweb View Post
I'd say 1:1 is as far as you'd go. Good data still holds at that level. I found that if the seeing is good and you dithered doing a drizzle integration will give you a final with details that are still sharp slightly past your image scale so close to 2:1. A good image will look ok at 1:1 and any res lower. Then there's the gear used as well. So although an image might have still edge aberrations if on-axis is excellent then I still see it as an excellent capture because the imperfections are limitations of the gear, not processing induced. Noise, reflections or chromatic aberrations don't bother me the least.
Thanks for your response Marc. It is good to see where different people land on this.
Reply With Quote
  #6  
Old 15-06-2022, 04:55 PM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
Yeah, but what if you use Topaz magapixel AI and increase resolution as a result, (obviously), with even sharper edges?.

Quote:
Originally Posted by multiweb View Post
I'd say 1:1 is as far as you'd go. Good data still holds at that level. I found that if the seeing is good and you dithered doing a drizzle integration will give you a final with details that are still sharp slightly past your image scale so close to 2:1. A good image will look ok at 1:1 and any res lower. Then there's the gear used as well. So although an image might have still edge aberrations if on-axis is excellent then I still see it as an excellent capture because the imperfections are limitations of the gear, not processing induced. Noise, reflections or chromatic aberrations don't bother me the least.
Reply With Quote
  #7  
Old 15-06-2022, 05:14 PM
sunslayr (David)
Registered User

sunslayr is offline
 
Join Date: Nov 2018
Location: Perth
Posts: 209
200% right on those corners, gotta see who has worse curvature or coma than me.

Last edited by sunslayr; 15-06-2022 at 08:23 PM.
Reply With Quote
  #8  
Old 15-06-2022, 06:18 PM
Ryderscope's Avatar
Ryderscope (Rodney)
Registered User

Ryderscope is offline
 
Join Date: May 2006
Location: Glanmire, NSW
Posts: 2,168
Quote:
Originally Posted by sunslayr View Post
200% right on those corners, gotta see who has a worst curvature or coma than me.
Very good David, I like your approach . Having said that, this is probably a good example to demonstrate my point. If you cannot see the coma when displayed on, for example, a 27" 4K monitor, why does one have to zoom in to see it? I will admit that I do this myself and we all like to try and get the best possible field and data. But if it doesn't affect the viewing, maybe we are becoming a little too obsessive, yes/no?
Reply With Quote
  #9  
Old 15-06-2022, 06:23 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by Bassnut View Post
Yeah, but what if you use Topaz magapixel AI and increase resolution as a result, (obviously), with even sharper edges?.
I hear it even grows claws and puts feathers on M16.
Reply With Quote
  #10  
Old 29-06-2022, 07:21 AM
marc4darkskies's Avatar
marc4darkskies (Marcus)
Billions and Billions ...

marc4darkskies is offline
 
Join Date: Mar 2007
Location: Quialigo, NSW
Posts: 3,141
Very late to this party but may as well put my 2c in.

I pixel peep my own images while processing up to 3-400%. Mostly this is to inspect noise levels and to reveal obvious processing and stacking artefacts around stars (including elongation) and then to evaluate remediation steps. Star quality can make or break an image, even when viewed native resolution! Fine detail in galaxy shots also warrants close inspection just in case I have gone too far. This is especially true when using decon or other sharpening tools.

I do tend to do some peeping at other people's images, but only the ones that look first rate on initial inspection. This is more out of curiosity and comparison, so I rarely comment on my observations.
Reply With Quote
  #11  
Old 29-06-2022, 09:33 PM
Ryderscope's Avatar
Ryderscope (Rodney)
Registered User

Ryderscope is offline
 
Join Date: May 2006
Location: Glanmire, NSW
Posts: 2,168
Quote:
Originally Posted by marc4darkskies View Post
Very late to this party but may as well put my 2c in.

I pixel peep my own images while processing up to 3-400%. Mostly this is to inspect noise levels and to reveal obvious processing and stacking artefacts around stars (including elongation) and then to evaluate remediation steps. Star quality can make or break an image, even when viewed native resolution! Fine detail in galaxy shots also warrants close inspection just in case I have gone too far. This is especially true when using decon or other sharpening tools.

I do tend to do some peeping at other people's images, but only the ones that look first rate on initial inspection. This is more out of curiosity and comparison, so I rarely comment on my observations.
The interesting observation here Marcus is the capability of the observer to detect issues at native resolution which possibly don't become obvious until zooming in to around the 300% mark. The implication is that the average human eye can detect very fine grained detail at native resolution such that underlying technical issues will become apparent, even though it is not obvious as to what they may be. It is that feeling when looking at an image and thinking that something is not quite right and then zooming in and spotting the problem. Or maybe we just spend too much time staring at screens
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 10:18 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement