Battling with stretching etc.

[earth titan]

Hi there.

Last week I took an Andromeda image and I've been trying to make the most of it. I don't think there is too much more I can do with it but I'm interested to see if anyone else can get more from it and if so - how!

The image is as spat out by DSS and the large .tif file can be found here (still uploading)

Here is the best I can manage using Gimp..

Thanks in anticipation.

[Guest]

I'm just getting started myself and using GIMP like you - I find its working pretty well as I can't afford photoshop.

I found that I was having to over-process my images to get rid of the gradient in the background (even after a healthy dose of processing in nebulosity). Now, flats will of course remove that too but I haven't managed them yet. So I was losing data trying to chase the gradient out of the image and that's when I found this free product that does it in a snap

Molecular Expressions Microscopy Primer: Digital Imaging in Optical Microscopy - Background Subtraction Toolkit Download

I do some basic processing in nebulosity > export to tif > open in GIMP > do a bit more levels adjustment > safe to jpeg > open in the app above and remove the gradient then complete processing in GIMP.

I can hear all the competent imagers rolling around laughing at that process as going to jpeg kills all kinds of data - but it seems to work for me :-)

Hope that helps - your image is looking good - and tonight will be clear in the UK!

David

[SJJ]

I can't access the tif file but looking at the jpg in Photoshop (and i'm by no means an expert on processing!) it looks as though the black point is clipped a bit so you may have lost some faint detail there.

[Jessun]

earth titan as I've said before this is an image I'd like to see less processed to be able to work out what's in there. After that the trick is simple but boring... add more subs. And then a few more. A very nice M31 I saw here had 9 hours total exposure.

[earth titan]

Hi folks.

Managed to post the link:

http://dl.dropbox.com/u/40788231/Post%20for%20SGL.TIF

Should work I hope.

I know it is only a few subs, but would be eternally grateful if someone could have a go. I'll just know if I'm getting somewhere then with the processing. Pretty please?

Ta.

[Carl M]

Here is my quick attempt..seems quite noisy. How many subs were there altogether? I think as stated above you are clipping the black point of the image which is getting rid of the noise but also elimanating part of M31.

[Spikey]

The link still doesn't work.

[earth titan]

Have posted a new link further down.

Thanks for trying.

When you say 'clipping the black point' I assume you mean when using levels. Is there a way of avoiding doing so?

[Carl M]

Yes, in levels when moving the black slider you should leave it just before it hits the foot of where the jump in data starts. Going over that point means that the data is being clipped and lost through processing.

[SJJ]

I've had a quick go with the cropped tif file in photoshop.... I think you have done a pretty good job with your image, maybe just washed out the galaxy core a bit too much.

[earth titan]

Thanks folks.

Looks like I'm not to far out then!

Need more subs but with the current scope (80mm) I'm happy so far.

[Spikey]

I've done a quick rendition of my own on the supplied tif. Looks like they all come out fairly similar.

Haven't done anything fancy like layer masks, this is just levels & curves plus some de-noising.

It's pretty noisy so a bit difficult to do much with only one image. To start with gather a load more subs plus some darks and stack again in DSS.

[earth titan]

Getting there.

Thanks for the help so far.

After a lot of work, I'm seeing more detail. Being more gentle with the processing.

Still a way to go, but no so bad

[ollypenrice]

It still looks black clipped and the colour balance is out. The top left of the histogram peak should be in the same position, as a rule of thumb, for eachcolour and you should allow a little flat to remain on the left of the histogram before the main peak stars to rise. As said above, you cut both noise and signal when you cut back too hard. You just need a lot more data. Imaging is about time. I rarely post anything with less than 10 hours data.

Olly