Help processing first dso's

[SlipperySquid]

So I'm loving this astrophotography malarky and wonder if anyone can help me improve the images I have?

I have quite a bit of noise especialy on the M81/82 image and a bit of vignetting which I would like to remove while still keeping detail!

I know in the long run it's going to be longer subs which I will be able to get once I sort out a guide cam etc, so I'm stuck with shorter subs.

If anyone would like to have a go at having a tinkle with these lower quality jpegs I would be most appreciative.

The progs I have is CS4 portable, Maxim and DSS.

Thanks in advance,

Matt.

[jgs001]

They're looking good. Flats at imaging time will deal with the vignette the best, one of the tools like Russel Cromans Gradient xTerminator will do a pretty good job though. Follow the tutorials on the site. I don't have any of the tools to do that sort of thing on this machine.

[Psychobilly]

Matt some of the actions need to be aplied to the images earlier in the workflow to get the best results...

Any where you can upload the tiff's out of DSS .. it might be an idea to reduce them to 25% and 16 bit (if they are 32 bit) size before uploading them

Peter...

[steppenwolf]

John has mentioned Russel Cromans' Gradient Xterminator and it is an excellent tool although as usual, there is no substitute for a good 'FLAT' as he has suggested to alleviate vignetting. However, this is what Grad. X and a little curves tweaking can do for one of your images:-

[roundycat]

looks to me as if you are well on the way if these are unguided images. If you want to learn PS there is no substitute for practice and there are some pictures on my web site that you can practice on.

Maxim can do all the pre-processing work such as calibration, alignment and combining into colour. When you save as a 16 bit TIFF to take it into PS Maxim does the maths in IEEE floating point but PS will effectively open it and definitely save it as a 15 bit image so don't worry about having squillions of bits in your initial image, once into PS it's outside your control.

Forget DSS.

Dennis

[SlipperySquid]

Thanks fellas for your input! I do use darks and bias frames but have yet to make a light box for flats, mainly because I'm to lazy! But with these cloudy nights and lack of darkness I ahve no excuses.

I'll try and redo the stacking in maxim and send you the files Peter! many thanks!

I'll also have a look at Grad x (Thanks John) and thanks for ahving a tweak Steve! where did all those extra stars come from!

I'll try stacking with Maxim but it seems more memory intensive then DSS but I'll have a go Dennis.

I am having wierd things happen when exporting images from maxim into PS or just saving them! Sometimes they come out very oversaturated but I think that maybe because I use the screen stretch? (I don't think it saves this setting when exporting an image?)

Thanks again fellas

Matt.

[roundycat]

don't use the Maxim screen stretch, or any other come to that. Save your work as a 16bit TIFF and go straight to PS and start stretching with Levels, there should be no problem.

Dennis

[steppenwolf]

I do stretch my files in MaximDL so that I can see what I've captured but am very careful not to clip the black point.

Maxim has two types of stretch, 'Screen' and 'Permanent'. The screen stretch is just that, it changes the file on screen only and is non-destructive. I agree completely with Dennis that you should save your work as a 16bit TIFF file and move onto PS A.S.A.P. I select 'File Filter' TIFF - 'Size Format' 16 bit - then click on 'Stretch' and select 'Permanent Stretch Type' linear - 'Input Range' Screen Stretch - 'Output Range' 16 bit.

This saves the file in 16 bit TIFF format with the 'screen stretch' applied so that I immediately have a recognisable image when I load it into PS.

[roundycat]

Steve, a question for you. When you apply a normal levels type of screen stretch in Maxim (as I do millions of times) it shows you roughly what is there. It never shows you a controlled contrast stretch from black to white as there is too much data to display.

If you set some kind of stretch in Maxim so as to have a recognisable image the instant you open it in PS, how do you know what damage it has done?

I never set a stretch in this way. I open the 16bit TIFF in PS and it is totally black except for a few bright stars. I then carefully stretch (gamma 2) using Levels until I can start to see the detail, usually three iterations does it. At no time do I move the white point. I then go to Curves for the fine detail work, stretching the image only where I want it to be stretched by careful placement of the adjustment points.

Dennis

[steppenwolf]

If you set some kind of stretch in Maxim so as to have a recognisable image the instant you open it in PS, how do you know what damage it has done?

Hi Dennis, I guess that any kind of stretch is potentially 'damaging' as it redistributes the levels within the data but I am not aware of the Maxim levels adjustment being any more or less damaging than that of PS.

What I have found by not doing a preliminary stretch in Maxim is that I end up having to do at least two stretches in PS to get a workable image to make an informed decision on for the next steps.

From memory, there is no difference in the final image whether I do a preliminary stretch in Maxim or not (it was some time ago that I stopped exporting the data to a TIFF without a preliminary stretch) but if you have some convincing evidence that there is an advantage in not doing a Maxim stretch, I'm listening!

[martin_h]

Had a bit of a play with one of them - cleaned up the background and tweeked a bit in curves

[roundycat]

Steve, you are right about the need to do an 'extra' stretch in PS before you can see a ghost of an image if you open it as an 'unadulterated' TIFF straight from Maxim. I invoke Levels, gamma 2, ok; Levels, gamma 2, ok; That takes about three seconds so I'm not bothered about the extra processing time!

What worries me about Maxim, and I freely admit that I have not explored it, is that the kind of arbitrary screen stretch that is often suggested might act in the same way as PS when running on auto. If you invoke curves and then auto have a look at the individual channel curves. They could all be shifted at the top and the bottom making a straight line and moving the white (and black) point in every channel. It is this unwanted shifting of the white point that puts me off doing anything that I can't actually see happening.

If you are actually suggesting using Maxim's equivalent of Curves that's a different issue.

Dennis

Edit; I just did both methods with a picture of 1396 in Ha. Using the Maxim stretch as advised and then opening in PS I compared it with a non-Maxim stretched version which I then stretched in PS using three iterations of Levels and one of Curves to try to match the PS version to the Maxim version. Without doubt the stars in the Maxim pre-stretched version were brighter, by quite a margin. This points to an unwanted and uncontrollable shift of the white point that ruins the subsequent processing if you are to retain star colour and prevent bloating.

Dennis