Is there a way of removing field curvature in Photoshop?

[michaelmorris]

My present imaging set up produces a degree of field curvature. What I need is a field flattener. The trouble is that equals money and at the moment I don’t have as much I want for buying astro stuff. (do I ever?)

Looking at the effects of field curvature on the image it strikes me that this looks very similar to star trailing, except that the direction of trailing varies with angle and the amount of trailing varies with distances from the centre.

Now, I know that there are several different ways of removing star trails in Photoshop.

http://www.focusmagic.com/

http://www.sightsabove.com/article.php?article=20

http://stargazerslounge.com/imaging-tips-tricks-techniques/113075-startrail-remover-ps-action.html

and probably many more I don’t know about.

So, is there a way of applying any of any star trail removal techniques so that the degree of star trail removal varies with distance from the centre and the angle of trailing removal rotates around the image?

[roundycat]

I'm certain the short answer is no. You might try playing with Filter Distort Spherize.

Dennis

[Ben Ritchie]

There's also a lens correction filter in CS2 (I think, going from memory so might be CS3) but I don't believe it's particularly good at correcting radial distortion of stars - Google might prove me wrong

[GME]

Sounds like using "Flats" might do the trick, you can read about it here DeepSkyStacker - Free click FAQ and pic the appropiate sub chapter

[Psychobilly]

Flats correct the uneven illumination (Vignetting) not field curvature...

[Sparrow]

Thank you, thank you, thank you. It never occurred to me and I've had to put up with coma in my images since I started my DSLR work (can't afford a coma corrector).

CS3 - yes the lense correction filter (filter/distort/lense correction) works brilliantly. I set the top slider in the dialogue box to -23 for my images. See the difference in the two images of the Pacman below - before and after:-

[ollypenrice]

Wow, that's pretty good!

Olly

[Sparrow]

Actually - having just rushed through all this year's images applying this filter, it seems to be about -15 on the slider setting to get the desired result for me - obviously it depneds on what scope and what camera etc as to which setting would work best - in my case it's the results from a Celestron Newt 8" F5 with the Canon 1000D DSLR

[Psychobilly]

Good Stuff will have to give it a go...

Worst case you will have to just crop the result slightly to get of the "corners"...

CS5 takes it a step further you can take images a series of test charts (how practical this would be with a scope is another matter) and it will then characterise the lens for distortion and light fall off (provided the target is evenly illuminated)... it also uses "content aware fill" to try and patch up the edges of the image....

http://labs.adobe.com/technologies/lensprofile_creator/

Peter...

[beamish]

Another great PS tip

[simon84]

Just had a play with it using the above image and it works so well. Really impressed with it. No flattener for me anymore......

[Ben Ritchie]

Glad it works, not sure where I got the idea that it didn't!

[Shibby]

Wow! I didn't know about this, the effect looks impressive - I must try it on my images!

[Sparrow]

Psychobilly - CS5 sounds great - beyond my pocket I'm afraid but I think this is really going to move my images on a stage. It really does begin to question of whether it's worth using a flattener or coma corrector at all - less glass to image through is always a good thing.

[Psychobilly]

Been thinkign about this a bit... more things to watch out for really....

although only time and trails will tell...

I uess that teh software processign is going assum ethe lens is producign a symetrical image with the same distortion in each "corner" ... so it will be in terestign to see what happens with an image wher ethen camera isnt quite square to the optical axis ...

I definitely dont see it as the "end" of the flattner/reducer as the latter also increases the FOV and "speeds" up the scope...

Peter...