Orpington Astronomical Society

Astronomy => Technical => Topic started by: RobertM on Oct 15, 2009, 18:47:51

Title: Diffraction spikes, merged...
Post by: RobertM on Oct 15, 2009, 18:47:51
Expect a whole load of new images with artificial spikes - Noel Carboni of 'Astronomy Tools' for PS fame has a new product ...

Star Spikes Pro

From the blurb:
QuoteIn response to popular demand for more control and realism in the generation of star diffraction spikes, we have just completed a year-long effort to take spikes to the next level.  Actually we jumped about four levels!

I'm trying to guess why he called it 'Pro' rather than 'Am' but other then revenue generation I can't think of much!  I'm sure the Pro's would rather not have them  :-?
Title: Re: Oh no, not more star spikes!!!!!
Post by: Rick on Oct 15, 2009, 20:03:00
"Pro" is usually short for "Probably costs too much"...

Quote from: RobertM on Oct 15, 2009, 18:47:51I'm sure the Pro's would rather not have them  :-?

If you search the academic astronomical literature you'll certainly find a whole lot more about trying to get rid of spikes...
Title: Re: Oh no, not more star spikes!!!!!
Post by: mickw on Oct 15, 2009, 20:23:53
OK,
I'm new to all this imaging malarky.....................
If I manage to image a great big white cross does NC have a tool to stick a star in the middle of it ?

:lol: :lol: :lol:

It's a shame really that someone with a real talent for teasing out extra detail from a subject should even want to add something fake.
Title: Re: Oh no, not more star spikes!!!!!
Post by: JohnP on Oct 15, 2009, 20:59:03
Well I've said it before & I'll say it again I'm not a star diffraction supporter but after seeing the results Noel got on my m45 image using the said 'Star Spikes Pro' tool I have to say it turned my mundane m45 image into something amazing (at least in my eyes & yes I know you wont all agree).. anyway that said I wouldn't spend $60 bucks on it... Now his set of Astronomy Tools actions is fantastic & for $19 they paid there way a 100 times over  the 4 or 5 years I have been using them. They also include a much 'coarser' star diffraction set of actions which I guess are the amateur ones verses his fully functional tweak everything & anything pro ones...

John.

PS - When I was into 35mm film photography I remember buying sets of Cokin filters to add diffraction spikes & other effects (& I bet I'm not the only one...!!!!).
Title: Re: Oh no, not more star spikes!!!!!
Post by: Rick on Oct 15, 2009, 21:07:37
Quote from: JohnP on Oct 15, 2009, 20:59:03Cokin filters
...for an artistic effect, not a scientific record.
Title: Re: Oh no, not more star spikes!!!!!
Post by: RobertM on Oct 15, 2009, 22:27:21
QuoteIf I manage to image a great big white cross does NC have a tool to stick a star in the middle of it ?

No, afraid not.  But there are tools in PS that you can use to get that 'starry' effect.
Title: Re: Oh no, not more star spikes!!!!!
Post by: Tony G on Oct 15, 2009, 22:50:49
Quote from: Rick on Oct 15, 2009, 21:07:37
Quote from: JohnP on Oct 15, 2009, 20:59:03Cokin filters
...for an artistic effect, not a scientific record.
But I thought the majority of the images taken by the imagers in our society, were processed more for the artistic look, than the scientific aspect.

Tony G
Title: Re: Oh no, not more star spikes!!!!!
Post by: RobertM on Oct 16, 2009, 06:58:51
Where did you get that strange idea from  :-?
Title: Re: Oh no, not more star spikes!!!!!
Post by: Tony G on Oct 16, 2009, 07:43:35
From all the pretty colours and diffraction spikes.  :roll: ;)

Tony G
Title: Re: Oh no, not more star spikes!!!!!
Post by: JohnP on Oct 16, 2009, 09:10:08
Quote...for an artistic effect, not a scientific record.

Exactly Rick !!!! - That's what I take images for if I wanted to take images for scientific effect I would just take single B&W raws.. I want images that look good hanging in a frame...

John.


PS - I keep all my raw data so if at any time I do feel the need to do some science on them no probs...
Title: Re: Oh no, not more star spikes!!!!!
Post by: Rick on Oct 16, 2009, 11:54:26
Quote from: JohnP on Oct 16, 2009, 09:10:08I want images that look good hanging in a frame...
...and if I was to do that I'd want images that were as real as I could get them, not fudged with fake effects. These days I look at stuff credited to Noel Carboni, and my first thought isn't "That's a cracking photo!", it's "What's real and what's faked in this?". The way he's going, it'll soon be nearer "Is any of that real?"...
Quote from: JohnP on Oct 16, 2009, 09:10:08PS - I keep all my raw data so if at any time I do feel the need to do some science on them no probs...
Glad to hear it.
Title: Re: Oh no, not more star spikes!!!!!
Post by: JohnP on Oct 16, 2009, 13:39:52
So Rick are things like...

non-linear stretches
noise reduction algorithms
unsharp masking
DDP algorithms
deconvolution filters
false colour images...
contrast enhancement
etc... etc...

classed as Real or Fake. Anyone of the imagers amonst OAS will tell you that getting data is only 10% of the process the other 90% is time spent processing & manipulating it into a good looking image.

My immediate thoughts whenever I look at any of Noel's images is OMG why can't I produce images like that...

John.

Title: Re: Oh no, not more star spikes!!!!!
Post by: Rick on Oct 16, 2009, 15:30:41
Quote from: JohnP on Oct 16, 2009, 13:39:52
So Rick are things like...
Ask yourself how the process is affecting the image:

For example:
non-linear stretch - used to adjust for distortion in the optical apperatus.
Adding anything? Yes, where pixels in the original are moved apart.
Subtracting anything? Yes, where pixels in the original are moved closer together.
Pixel values will be adjusted using an appropriate formula to avoid introducing artefacts.
Justifiable? Within reason. Too severe and the artefacts will ruin the data.

Noise-reduction algorithms
Adding anything? Hopefully not.
Subtracting anything? Hopefully, only noise.
The effectiveness of any particular algorithm will depend on how well it actually identifies noise.
Justifiable? Probably, provided the noise identification algorithm is effective. Needs to be treated with a bit of caution, though.

Unsharp masking is a relative contrast adjustment process. Applied properly it will reveal real detail. May also reveal noise... Likewise for plain contrast enhancement.

Standard flats, darks, etc. are attempting to measure and correct for distortions (in amplitude, mostly) of the sensor, and there's plenty of scientific literature explaining and justifying each step of the process.

Other algorithms have to be evaluated one by one. A good process will reveal as much real detail as possible while introducing as few artefacts as possible. Deconvolution is one such.

False colour is a recognised presentation technique, and justifiable so long as the mapping is declared.

etc...

We've had this discussion before. I consider the adding of fake information (whether that's fake diffraction spikes, fake nebulosity, fake stars, or fake flying saucers) to an astrophotograph to be deceitful and unacceptable. You aren't going to change my mind on that. Diffraction spikes are part of the information in an astrophotograph. Genuine spikes tell us something about the object being imaged and about the optical apperatus used to obtain the image. Fake spikes lie to us about the object being imaged and about the optical apperatus used to obtain the image.

Conversation over.
Title: no option to reply
Post by: Fay on Oct 16, 2009, 16:17:25
Rick, I am not given the option to reply on "Oh not more star spikes" why is this please? All other posts ....ok
Title: Re: no option to reply
Post by: RobertM on Oct 16, 2009, 16:35:01
Blame the forum :police:
Title: Re: no option to reply
Post by: Rick on Oct 16, 2009, 17:09:09
The conversation's going round in circles. We've had it before. It was accrimonious then. It's turning accrimonious now. Time to stop.
Title: Re: no option to reply
Post by: Fay on Oct 16, 2009, 17:16:20

Rick, my query is a genuine one, only when going into that post I get the following options in this order: Notify, Mark Unread, Print
Title: Re: no option to reply
Post by: Mac on Oct 16, 2009, 17:23:32
?
Title: Re: no option to reply
Post by: Mike on Oct 16, 2009, 18:38:19
 :o
Title: Re: no option to reply
Post by: JohnP on Oct 16, 2009, 20:11:32
Rick think your out of order saying your piece & then closing the discussion... can't really see the point in contributing to forum discussions if I can't have my say....!!!!!

Cya - John
Title: Re: no option to reply
Post by: JohnP on Oct 16, 2009, 20:23:23
Quoteaccrimonious
..... ?
Title: Re: no option to reply
Post by: Mike on Oct 16, 2009, 20:56:26
I have to say I agree. We should be free to say what we want even if some of us don't like it. As long as what is being said is not abusive.
Title: Re: no option to reply
Post by: Fay on Oct 16, 2009, 22:07:21
Mike, i was about to say, how expensive the new star spike program was by Carboni, but could not get REPLY up for this post
Title: Re: no option to reply
Post by: Mac on Oct 16, 2009, 23:06:39
you could always cut and paste in to this thread and carry it on from there.
or restart a new one.
Title: Re: no option to reply
Post by: Rick on Oct 17, 2009, 10:58:08
Quote from: Fay on Oct 16, 2009, 17:16:20Rick, my query is a genuine one, only when going into that post I get the following options in this order: Notify, Mark Unread, Print
The diffraction spikes thread is locked. If you have a query about something else you're welcome to start a new thread.
Quote from: JohnP on Oct 16, 2009, 20:11:32
Rick think your out of order saying your piece & then closing the discussion... can't really see the point in contributing to forum discussions if I can't have my say....!!!!!
We've had the discussion before. You're raising a technical argument against something I see as a moral integrity issue. I can't see any point continuing the discussion, but if you really want to have a last word added, please PM it to Ian or me...
Title: Re: Diffraction spikes, merged...
Post by: Ian on Oct 18, 2009, 16:59:13
Right, I've had a think about this, and this is what I'm going to do.

This topic will be unlocked tomorrow at 9am, for 24 hours only. Anyone is then welcome to post one reply stating their final and binding position on diffraction spikes. Multiple posts by the same person will be deleted, as will anything that is not regarding your personal views on diffraction spikes.

The topic will then be locked and stickied so the next time anyone wants to discuss diffraction spikes, all we need do is look up their views stated here.

It's either that or I introduce a naughty step.

For the record, I don't give a toss about diffraction spikes.
Title: Re: Diffraction spikes, merged...
Post by: mickw on Oct 19, 2009, 10:53:11
I think the choice to add diffraction spikes should be left to the individual that produced the image, and not a rule imposed by somebodys moral code.
Title: Re: Diffraction spikes, merged...
Post by: Carole on Oct 19, 2009, 18:12:29
QuoteI think the choice to add diffraction spikes should be left to the individual that produced the image, and not a rule imposed by somebodys moral code.
Yes I'd agree with that.

Personally I like diffraction spikes but only if they are not put on afterwards artificially. 
Therefore my opinion is:
BRING OUT THE BEST OF WHAT YOU HAVE CAPTURED, SO LONG AS NOTHING ARTIFICIAL IS ADDED.
Having said that though, John's M45 is fabulous to my eye, spikes and all. 

Carole