How to save in PI in a way that imports into PS well

[Mark de Regt]

Hi,

I know it's heresy, but there are things I can do more easily (or at all) in Photoshop that I cannot (or at least cannot yet) do in PI.  For instance, when a very bright star, whether in or out of the FOV, casts a huge light umbrella (as very bright blue stars can do), it's pretty easy to tame it in Photoshop, but I have no idea how to tame it in PI.

Anyway, when I save the image in PI, whether I save as a TIFF or FITS, importing the saved file into PS brings up a cloudy image.  I can pretty much (but not totally) fix it with "levels," but I am hoping I'm just missing something, and there's an easy way to do it.

Thanks for any helpful suggestions.

Mark

[Zocky]

You should save file as 16 bit TIFF.

[Mark de Regt]

I do; still have the problem that it doesn't open as a nice clean file like the one I saved: I have to mess with the sliders in PS, and even then it's not quite the same.

[oldwexi]

Hi Mark!
PI usually saves the data normalized, with Floating Point numbers between 0.000000000 and 1.0000000000 !
If you want to save the images for painting programs (why?) you have to select
TIF and 16-Bit Integer(!!)
Than the drawing programs can also read the images.

I remember some years ago i did also the image painting and software jumping  and as a result i did not get any progress with PI.
So i removed ALL painting programs from my computers. From that day on the progress
to understand how to use PI increased significantely. Today for me PI covers all image processing needs.

May be we can help you to do in PI what you think you cant do wit PI.
Let us know in details what you do in the Painting Software (PS) when you "tame" a star umbrella

Gerald

[Mark de Regt]

May be we can help you to do in PI what you think you cant do wit PI.
Let us know in details what you do in the Painting Software (PS) when you "tame" a star umbrella

Gerald

Thanks; I'll try this.

The thing that first comes to mind which I have no idea how to deal with in PI:  When I have a very bright (usually blue) star, either in the FOV or just outside the FOV, it can create a huge blue disc around it, decreasing in brightness as the distance from the offending star increases; it's not reflections, but just a huge blue glow.  In Photoshop, I can reduce that to a reasonable size easily; I have no idea how to begin doing that in PI.

Also, I save my images in various sizes before posting on the Web; that's easy in PS, but I have no idea how to do that in PI.

Thanks!

Mark

[Geoff]

Of course the real solution is not to try a software fix at all. Here is a quote from Juan in relation to fixing tracking problems, but it applies equally to your problem:

So the only solution to tracking problems is fixing the hardware that is causing them: throw away all wrongly tracked frames, look for the causes of bad tracking in your equipment, analyze and understand them, and fix them. Then acquire the data again. Astrophotography requires the pursuit of excellence, both technically and aesthetically. No software 'magical box' can help you to achieve that.

Geoff

[sreilly]

Geoff,

This is not a tracking problem as you noted but I don't see what possibly needs fixing other than maybe not imaging this target if you can't frame it away from the offending star. Seeing how you can't move the star I guess you just get a more narrow view but then maybe it's not captured at all or very little. When a very bright star is just off the field of view of the imaging chip it certainly casts its light still and can have a strong influence on the captured data. The Flame Nebula comes to mind. There is no equipment to fix and there is a problem with the glow. I also had this issue with M45 using my 12.5" RC. You could say to use a different telescope with a wider view to maybe tame it a bit but then that doesn't address the issue. Seems I had this problem and someone guided my through using DBE to eliminate or very much tame the effect and it worked well. That's the kind of answer Mark is looking for, of course with some useful information on how to do this.

Mark, a suggestion, can you post this image to show the problem? And yes, I do understand the temptation to go back to a program that you know how to take care of these problems but if you're like me you would like to find out how to do these things using PI.

As for saving for web sizing, use Resample and input your desired size.

Steve
 

[llpastro]

Dave's Astrophotography blog has a tutorial on halo removal that might help.  The address is http://trappedphotons.com/blog/?cat=5

Larry

[RickS]

There is a StarHaloReducer script but I never had much success with it.  I usually just build a mask and use a range of tools like CurvesTransformation, HistogramTransformation and MorphologicalTransformation to reduce the halo.  It can take a few iterations to do a good job on a big halo and requires some facility and creativity with PixelMath.

Cheers,
Rick.

[Geoff]

Geoff,

This is not a tracking problem as you noted but I don't see what possibly needs fixing other than maybe not imaging this target if you can't frame it away from the offending star.
Steve

Maybe baffling and/or flocking the optical path could help.

[Mark de Regt]

They're all just tools, aren't they?  If I don't know how to solve a problem in one tool, but I can solve it in another, I'm going to do that.

The bright stars aren't a matter of flocking, or technique, or any other fault; they just are; a mag 3 blue star in a 15-minute sub is going to leave a significant footprint.  It's not a "halo," nor is it a reflection.  But I can make it go away in Photoshop.  If I can make them not be a problem in PS, I'm going to use PS.  If there's a way to do that in PI, I'm all ears, but I'm not going to stop fixing problems just because I don't know how to solve them with one tool, when I know how to solve them with another.

[Warhen]

Mark, saving FITS to 16-bit TIFF does/should match-up precisely between PI's and PS's visualization. Read David Ault's excellent tutorial http://trappedphotons.com/blog/?p=616 on halo removal. Gerald, Photoshop is the de facto standard of professional graphics compositing tools. :>) Yes, one can also paint with it (I don't consider judicious refining of masks painting either).

[oldwexi]

Hi Mark!
Saving jpeg in different sizes in PI you can do with the script "WriteJPEGWithSizeConstraint" startable like any
PI process from the Script tab.

Concerning using different Software Tools -
The difficulties start with Software Jumping.
The more or less only exchangable format between professional Astrotools like PI and a
professional painting program like PS  is 16-Bit integer. So, compressing a 32-Bit stacked image
to 16-Bit reduces already dynamic. So moving a 32-Bit floating point image to another SW which cannot interpret it
therefore is not what i recommend to do, if you want to keep finest details.

Therefore i decided to stay in the 32-Bit world and stopped sofware jumping.
in addition to data loss , the more different software you use the more you have to learn different Software Tools and interfaces,
it will be difficult to raise the knowledge on all sw tools to the same high level.

I would really like to see described the steps you use for removing the "umbrella" in the PS.
If it is image processing and not handpainting i am sure PI can do it also...

Gerald

[Mark de Regt]

Thanks, Warren.

As to how I get rid of the blue umbrellas around bright stars in PS....

It's pretty simple, if painstaking.  And I'm sure some people wouldn't approve, but I don't do this for them, so I really don't much care. 

I select the blue haloish region, putting a very significant feather on it (really not much different from a mask in PI, but a heluva lot easier for me).  I very gently lower the saturation from the color layer, and the brightness (usually using levels) from the luminance layer.  I repeat this process, a tiny bit at a time, making the selection a bit smaller each time, until the umbrella is a size and intensity I'm happy with .  It takes time, and sometimes I have to start over, but it works well (especially if there aren't any parts of the real object in the umbrella; that can complicate matters).

[jkmorse]

Mark,

All I can say is: Blasphemy!! 

Best,

Jim