How far is it acceptable to reduce stars? [Deep Sky] Processing techniques · Carastro · ... · 97 · 5481 · 4

carastro 8.04
...
· 
·  12 likes
I see so many images these days where the stars have been reduced so much that they are in some cases almost non existent.  

l feel the image has then ceased to be an astronomy image and become more a piece of artwork. 

l also fear this is becoming the NORM.  

Any one else think the same?  
Been bugging me for a while.
Edited ...
Like
Starman609 6.45
...
· 
·  2 likes
Great topic for discussion. I agree with you 100%. I also have seen images where the stars are barely there to the point where it is noticeable. I'm guilty of reducing the star brightness/size too but only to the point where they don't overshadow the object I'm trying to photograph. My personal preference is against bloated stars as I feel they reduce the beauty of the image and distract from what I'm trying to capture.

Here is one of my latest images, feedback welcome.

https://www.astrobin.com/joksd5/E/

CS, Eddie
Edited ...
Like
andreatax 7.90
...
· 
·  1 like
Well, if all you do it NB, can you actually complain about stars size or lack thereof?

Mine, they are all very starry.
Like
carastro 8.04
...
· 
·  2 likes
I feel a bit of star reduction to show up the nebula better is acceptable but l feel in some cases this has been overdone.
Like
andreatax 7.90
...
· 
·  1 like
The way I see it doing strict NB imagery is just as bad as heavily reducing/dimming stars.
Like
apophis 0.90
...
· 
·  1 like
andrea tasselli:
Well, if all you do it NB, can you actually complain about stars size or lack thereof?

Mine, they are all very starry.

Thats not so as i mostly do NB and as here stars are textured and colorful.
https://astrob.in/5t32zn/0/
Roger
Like
apophis 0.90
...
· 
andrea tasselli:
The way I see it doing strict NB imagery is just as bad as heavily reducing/dimming stars.

??
Roger
Like
andreatax 7.90
...
· 
Roger Redcat:
Thats not so as i mostly do NB and as here stars are textured and colorful.
https://astrob.in/5t32zn/0/
Roger


They seem to me pretty monochrome, same color of their background in fact. Most definitely not very colorful.
Like
apophis 0.90
...
· 
andrea tasselli:
Roger Redcat:
Thats not so as i mostly do NB and as here stars are textured and colorful.
https://astrob.in/5t32zn/0/
Roger


They seem to me pretty monochrome, same color of their background in fact. Most definitely not very colorful.

I didnt take any off though. And the nebula is the target not the stars but still would never do a starless image.
Roger
Edited ...
Like
andreatax 7.90
...
· 
Roger Redcat:
I didnt take any off though.
Roger


Still way less than there should be. Way, way less.

In truth my comments applied to the OP complaints, as it were.
Like
apophis 0.90
...
· 
andrea tasselli:
Roger Redcat:
I didnt take any off though.
Roger


Still way less than there should be. Way, way less.

In truth my comments applied to the OP complaints, as it were.

How many do you want , 10 , 20 , 30 million. As i said the target is the target ,the stars are consequencial but necessary.
Like
carastro 8.04
...
· 
·  1 like
andrea tasselli:
The way I see it doing strict NB imagery is just as bad as heavily reducing/dimming stars.

***Unfortunately some of us have no other option when we live in heavily LP locations.   But what l am referring to also applies to heavily star reduced NB images.
Like
PathIntegral 5.01
...
· 
·  15 likes
Even for scientific purposes, stars sometimes are mere distractions in the foreground. As data stars and other content are definitely separable most of the time (except those in the star formation regions). Also, the true sizes of stars in an image are almost zero and they can be regarded as point sources. The "size" of stars that appears in an image comes from seeing effects and optical diffraction, none of which are "astronomical" in nature. So in this sense I don't think star reduction (or elimination) is a matter of art, but rather that of data science and optics.

The issue is, however, star reduction has sometimes been done in a way that leaves a lot of artifacts that got in the way of true signals. This is when it becomes art, and not a great one at that. So my view is, as long as star reduction can be done without leaving artifacts and enhances the signals of interest, the more the merrier!
Like
andreatax 7.90
...
· 
·  1 like
Roger Redcat:
How many do you want , 10 , 20 , 30 million. As i said the target is the target ,the stars are consequencial but necessary.


As many as they are. And I would flip the correlation, the target is the stars the other is a consequence.
Like
andreatax 7.90
...
· 
Carastro:
Unfortunately some of us have no other option when we live in heavily LP locations.   But what l am referring to also applies to heavily star reduced NB images.


I suspect that you do it yourself, I mean heavily reducing/dimming the stars. Is it not?
Like
carastro 8.04
...
· 
No l don’t heavily reduce stars.  I slightly reduce stars.
Like
PathIntegral 5.01
...
· 
·  4 likes
Yuxuan:
Even for scientific purposes, stars sometimes are mere distractions in the foreground. As data stars and other content are definitely separable most of the time (except those in the star formation regions). Also, the true sizes of stars in an image are almost zero and they can be regarded as point sources. The "size" of stars that appears in an image comes from seeing effects and optical diffraction, none of which are "astronomical" in nature. So in this sense I don't think star reduction (or elimination) is a matter of art, but rather that of data science and optics.

The issue is, however, star reduction has sometimes been done in a way that leaves a lot of artifacts that got in the way of true signals. This is when it becomes art, and not a great one at that. So my view is, as long as star reduction can be done without leaving artifacts and enhances the signals of interest, the more the merrier!

In the same spirit, you may also ask how much background darkening is acceptable to hide noise. We know the true spatial background is almost completely dark, so in principle many images, especially those of galaxies, should have an absolutely black background. But the problem is that bluntly doing so often leads to artifacts which also truncates true signals.
Like
andreatax 7.90
...
· 
·  1 like
Carastro:
Unfortunately some of us have no other option when we live in heavily LP locations.   But what l am referring to also applies to heavily star reduced NB images.


I suspect that you do it yourself, I mean heavily reducing/dimming the stars. Is it not?
Yuxuan:
Even for scientific purposes, stars sometimes are mere distractions in the foreground. As data stars and other content are definitely separable most of the time (except those in the star formation regions). Also, the true sizes of stars in an image are almost zero and they can be regarded as point sources. The "size" of stars that appears in an image comes from seeing effects and optical diffraction, none of which are "astronomical" in nature. So in this sense I don't think star reduction (or elimination) is a matter of art, but rather that of data science and optics.

The issue is, however, star reduction has sometimes been done in a way that leaves a lot of artifacts that got in the way of true signals. This is when it becomes art, and not a great one at that. So my view is, as long as star reduction can be done without leaving artifacts and enhances the signals of interest, the more the merrier!

*** 

I couldn't disagree more. Diffraction and scattering are not mere by-products of imaging but a direct consequence of the entirety of the physical process of producing an image via a telescope. And they do NOT have zero size, just a very tiny one. If you want to get rid of them, be my guest. I can't care less.
Edited ...
Like
PathIntegral 5.01
...
· 
·  4 likes
Yuxuan:
Even for scientific purposes, stars sometimes are mere distractions in the foreground. As data stars and other content are definitely separable most of the time (except those in the star formation regions). Also, the true sizes of stars in an image are almost zero and they can be regarded as point sources. The "size" of stars that appears in an image comes from seeing effects and optical diffraction, none of which are "astronomical" in nature. So in this sense I don't think star reduction (or elimination) is a matter of art, but rather that of data science and optics.

The issue is, however, star reduction has sometimes been done in a way that leaves a lot of artifacts that got in the way of true signals. This is when it becomes art, and not a great one at that. So my view is, as long as star reduction can be done without leaving artifacts and enhances the signals of interest, the more the merrier!

*** 

I couldn't disagree more. Diffraction and scattering are not a mere by-products of imaging but a direct consequence of the entirety of the physical process of producing an image via a telescope. And they do NOT have zero size, just a very tiny one. If you want to get rid of them, be my guest. I can't care less.

Astronomically they have ~zero size compared to other objects. Period.

Saying you "can't care less" sounds rude and is completely uncalled for. At no point did I insinuate that it is something you should care for.
Edited ...
Like
andreatax 7.90
...
· 
·  1 like
Carastro:
No l don’t heavily reduce stars.  I slightly reduce stars.


Then you must do a very good job because they pretty much look starless to me.
Like
andreatax 7.90
...
· 
Yuxuan:
Astronomically they have ~zero size compared to other objects. Period.


They do not. Period.
Like
PathIntegral 5.01
...
· 
·  1 like
Carastro:
No l don’t heavily reduce stars.  I slightly reduce stars.

Well, keep in mind that scientifically, the colors and the sizes of stars in a NB image does not have much meaning (their coordinates do). Stars have continuous spectra and zero size. (By zero size here I mean their angular size is much much smaller than a pixel of your CMOS sensor).  So using their apparent sizes at three specific wavelengths to represent stars doesn't characterize their true astronomical nature either.
Edited ...
Like
PathIntegral 5.01
...
· 
·  1 like
andrea tasselli:
Yuxuan:
Astronomically they have ~zero size compared to other objects. Period.


They do not. Period.

Thanks for your opinion.
Like
Anderl 3.81
...
· 
·  7 likes
Just do it as you like it ;)
Like
mackiedlm 5.92
...
· 
·  10 likes
I definitely think there is a place for star reduction - both in BB and NB. There are images where the number and size of the stars stop one from seeing the full beauty and detail of the subject and I think thats the time for it. I do fully agree that it can be overdone and its needs to be done with a view to the overall image and it is sometimes difficult to strike the balance. I have sometimes looked back at an image and seen that I did indeed overdo star reduction and have gone back and brought them back up. (reduced the reduction?? )

So I think that really the answer to the original question can only be "It depends"!!

And like so much in astrophotography that has a lot to do with ones personal preferences. For me it is different for each image and is based on the premise that the stars should be reduced only as far as is needed for the subject matter to become the main draw of the image but not so much that they become insignificant (but that again is pure personal judgement!)

I am not a great fan of entirely starless images but there are some where I can appreciate the chance to really focus on fine detail of the subject - so long as the stars are not replaced by artifacts. But i always like to see a starred image alongside the starless, it puts the subject back into context.
Like
 
Register or login to create to post a reply.