may as well ask this question as its been on my mind a while
after stacking images and you open them in photoshop what is the rule of thumb for processing.
i know about levels , curves , sharpening etc.
but is there a a uniform order in which these or certain tools must be applied
Personally, I don't have a special particular order. I usually start by making a copy, then play with Curves (3 point), then 'Space Noise Reduction' in PS Actions, then I do a bit of Contrast/Brightness, then if I'm happy, I'll finish off with a bit of 'Sharpen' if I think it needs it. Other times I play with everything (including PS Actions), and if I don't like the result, I make another copy from the original and try again and again.
That is true Ken and i do the same really, but there has to be a sequence of events to get the moat out of an image, and to know when to stop, and not over process, just to have it look good isn't always the correct procedure, i think.
Don't worry mate, i do like you and play and play till it is properly buggered and then have an other go.
If this is the procedure that others use then i am on the right track i suppose, but was hoping for a more experienced approach from some other members.
Ken, and just to add to that , i to have tried that actions addon in photo shop, but have to admit i don't like it at all, and although Doug, was so kind and sent it to me, because i lost it, so to speak, i have actually put it on the shelf and probably wont use it again.
After initial processing, I usually go into photoshop and duplicate a couple of layers, this will allow me to adjust different depths without stuffing the image up. On one layer I adjust colour balance and try and bring out more detail in the Neb, using colour balance and saturation, as well as curve adjustment and levels. Another layer is used to adjust background darkness via curve adjustments again and levels. and another one to smooth out noise by applying a median filter and put this layer at the bottom. All layers once you are happy with them can be divided into their own percentage of opaicancy like 25-25-25-25 to make up 100%. This works with some images, but not all, you have to be a bit selective what images you apply it to. Once flattening the image you can do final adjustments in brightness and contrast.
I have had up to 10 layers on one image and they turn out okay. If you can get hold of a book for Astrophotography in Photoshop or something like that, it will be a great help. Again it's up to the individual and the image to the way you process.
If your unsure on curves adjustment, go into You Tube and type in the search Photoshop curves, it has some pretty cool tutorials.
I think a lot depends on what you want and what you perceive needs changing.
For example, an image with tracking errors resulting in egg shaped stars needs a bit of offset filtering before any curves and levels adjustments are made.
Curves should be used in conjunction with levels to ensure among otherthings that black point is not clipped (white point too)
Any nebulosity that you think would befefit by sharpening should be done using a Highpass filter technique rather than a blanket unsharp mask treatment that will mess the star images up.
Stars that are a bit bloted or washed out need selective treatment to bring out colour without saturating the rest of the image.
Probaqbly the best approach might be to submit an image with a statement of what you would like to see improved, and a request for info on how to do it; just asking for opinions will result in too many different ideas. That way you will learn how to do different things to get what YOU want. For example if you waned to improve the background noise in an image, but were not specific, you might be told that there is too much green or too much field curvature or not enough stretching or something.....anything but what you really want to know.
But to return to your original question,
Using a 1shot colour imager, you should
1. calibrate your raw image using dark frames and flat fields.
2. convert each calibrated sub frame to colour (debayer)
3. align and combine your subframes.
4. assess the resultant image, (colour should have been balances at step 2)
decide what needs fixing/enhancing and go from there, as needed.
I know many people stack before converting to colour, and many down load in colour automatically. but this is not the best method to use and for some sound reasons.
Doug
This is basically what I currently do. It is always changing as I learn more. It is not necessarily the best or only way.
I always use in camera noise reduction (ICNR) and generally expose at an ISO of 500 as the 5DH still has maximum dynamic range at this ISO setting. For a 300D or 350D etc an ISO of 200 gives you maximum dynamic range. The maximum dynamic range is determined by the A to D converter in the camera and is 12 bits. How high you can set ISO without losing dynamic range depends on the sensor. In the case of the 5DH the sensor has a potential dynamic range of 14.3 stops or bits.
1 RAW to TIFF using Images Plus
2 Correct for flats using IP
3 Digitally develop in IP Backgd Wt set at 0.65 adjust Break-Point so histogram is clear
of left hand side ie no faint data clipped
4 Stack with RegiStar and always use median averaged to minimize noise
5 Adjust RGB levels independently in Photo Shop and maybe a bit of brightness to get
background ‘correct’.
6 If I use curves I generally use the form in the image below.
Bert and Doug, thanks for your responses, that really helps a lot, although i don't do it exactly the way you have described it, i have to admit i'm pretty close, i thought it was much more involved in this procedure, but now i can see that i'm on the right track.
Thanks Again
Leon
Last edited by leon; 28-07-2007 at 11:26 AM.
Reason: spelling errors
thanks from to for your responses to our question , it has indeed shone some light onto the processing area of astrophotography and i now understand that every image needs a similar but yet different processing technique , just understanding layers is way hard to get to grips with , but i suppose once you have got the general jist of it , you can advance onto refining the layer technique's
many thanks
I think if you open an image in Photoshop and 'duplicate layer' a few times you will see that each successive layer defaults to a 'background copy'.
This means that PS duplicates from the original background each time, not the active layer.
There are other ways of making a layer, and then you have the option to use the previous layer or not.
Duplicating layers is not the only way to use layers, layers are very powerful tools and there are a lot of things to learn and remember.
It is possible to process a pleasing image without the use of layers. Best bet would be to follow up on the web tutorials and or obtain a book on using photoshop.
Try the auto levels, auto color, auto contrast on your first copy and see if you can match it, with second third and an infinite number of copies trying all the stuff available ... when you have about 100 pick one you like .
My point is experiment it is a lot of fun and you get to see the effects and analyse what is working for you at that stage of your journey.
alex
... when you have about 100 pick one you like
.
it is a lot of fun and you get to see the effects and analyse what is working for you at that stage of your journey.
