Bootstrapping....

[Zakalwe]

OK, here's the deal.

I am a total noob at this malarkay, and I have read a hundred tutorials on how to process images. I think I have improved over the last couple of months, but one of the biggest issues that I have is that I feel like I am "bootstrapping"...ie, trying to pull myself into the air by my own bootstraps.

I feel like i don't really have a solid foundation to start from, by that, I am not sure if my initial captures are good enough (garbage in, garbage out), what method I should use in DSS/PS/Registax etc in order to post-process the images. Finally, I often don't know what "good" looks like (ie colour balance, saturation etc), which comes from experience.

Looking at things like the PS workflow thread further adds to the confusion, as everyone seems to have a different way of doing things.

I think I would love to spend some time with someone more experienced (oooeeer missus:eek:) and see how they would process my images. Maybe they would throw away the base material (garbage in, garbage out again!). If I could see the steps, step-by-step, then I would have a little more knowledge on which of the million different techniques to apply!

Soooooo.... Is anyone interested in helping an eejit like me with some of my source material? It could make a decent tutorial for others too?

[Sparrow]

I do sympathise with you.

There is no one correct way to process images - different techniques work for different people, there are choices in what software to use for processing, and as every image is different, each image demands different techniques. I think what you need to do is adopt one of the more straight forward workflows, practise, and experiment. See what works best for you, read advice on here and try out different methods.

The golden rule is, don't throw away your raw data so that you can always start from fresh again with different techniques as you learn them

My work flow is roughly as follows for a typical Deep Sky object such as a nebula - this is not a step by step guide on how to actually do it:-

1. I go throu my subs manually and ditch any that I don't like.

2. Load my subs, darks and flat frames in DeepSKy Stacker and stack them

3. Open the resulting image in Adobe CS3

4. Convert to 16 bit

5. Adjust levels - usually only one or two iterations

6. Use curves to gradually bring out the nebula - better to make several small iterations instead of few large ones

7. Adjust brightness/contrast

8. Use Pixinsight to remove gradients (Dynamic Background Extraction)

9. Return to Adobe and make further final adjustments to curves

10. Smooth the image to remove grain.

This is only a very basic workflow and I am actually incorporating more steps now to reduce star size and smooth the image but it might give you a starting point in the meantime.

Hope this helps

Regards

John

[libraryman]

I find these tips helpfull John, with regard to subs and having recently taken a 1st image; i am judging the subs i take by the roundness or otherwise of the stars.

combined with apparent focus as i look at my subs i feel that if this basic critera is acheived then what ever info is in the data i can draw out at leisure.

would this veiw point be right?

Ray

[Sparrow]

Yes absolutely Ray - Once you've got the data you can play with it to your heart's content.

I started imaging 3 years ago and started on DSO work 2 years ago. I'm now going back on data from 18 months ago reprocessing it now that my techniques and knowledge has improved. The other thing is, that if you collect a certain amount of data in one year for a specific target, as long as you know what set up you were using (scope and camera etc) You can capture more data the next year to add to it.

DSS will check subs for quality and you can specify a %age for it to include, but I prefer to check through the subs myslelf for quality - as you say, I'm looking for roundness and sharpness of stars, along with the general SNR and any stray images with aeroplane lights or satellite trails in them.

I'm lucky that I've got an experineced "guru" friend who has given me loads of help along the way, but you can only digest so much at a time, so it;s best to start simple and then expand on it.

Regards

John

[Zakalwe]

Thanks for the responses guys, I appreciate it.

I'm lucky that I've got an experineced "guru" friend who has given me loads of help along the way, but you can only digest so much at a time, so it;s best to start simple and then expand on it.

Regards

John

Thats the bit i am struggling with.....I would love to see someone experienced playing with my data to see what can be extracted from it.

[Helen]

Where are you Z? We're planning on offering some one-one sessions with experienced imagers (eg MartinB, RobH) at SGL6, these will do exactly what you want ie take your data and work through it with you.

Helen

[Zakalwe]

Where are you Z? We're planning on offering some one-one sessions with experienced imagers (eg MartinB, RobH) at SGL6, these will do exactly what you want ie take your data and work through it with you.

Helen

Thanks for the reply Helen. I live near Lancaster, and am following the Northern Imaging Day thread with interest (though it's been live for nearly a year with no confirmed date or location?).

Something like that would be really, really useful:icon_salut:

[Sparrow]

You could always convert one of your raw stacks to jpg and post it on here and ask if people would have a go at processing - the only problem with that is that the compressed file inevitably loses some of the data and won't stand the processing as well. Alternatively if you get set up with something like MSN Skydrive, you could post files up to 50mb and provide a link to download them so you could post an uncompressed Tiff file as long as you keep the file size below 50mb.

[m.tweedy]

here's a previous thread that has quite a few tips and guidance

http://stargazerslounge.com/imaging-image-processing-help-techniques/121390-curves-levels-advice.html

[Zakalwe]

You could always convert one of your raw stacks to jpg and post it on here and ask if people would have a go at processing - the only problem with that is that the compressed file inevitably loses some of the data and won't stand the processing as well. Alternatively if you get set up with something like MSN Skydrive, you could post files up to 50mb and provide a link to download them so you could post an uncompressed Tiff file as long as you keep the file size below 50mb.

Thats sorta what I was hoping for TBH. But I want to see exactly the steps taken and the workflowto see how the results are achieved.

[roundycat]

see if there is anything here that helps.

Mono to RGB

Dennis

[Zakalwe]

Holy thread resurrection, batman:D

Right, if any kind soul out there wants to have a go and see if there is any usable data in here, I would be really grateful:

Click Me

(Big file- 145Mb TIF)

16 x 200 sec, 10 x 150 secs lights and 11 x 200 sec darks.

Unguided

Stacked in DSS.

Unmodified Canon 50D, SW Equinox 80mm Pro, SW field flattener and SW LP filter

Thanks in advance for any help offered:icon_salut:

[AndyUK]

First, thanks for referring me to this thread ...

Alternatively if you get set up with something like MSN Skydrive, you could post files up to 50mb and provide a link to download them so you could post an uncompressed Tiff file as long as you keep the file size below 50mb.

I use dropbox which has no specific single file upload restriction other than the size of the box itself - They give you 2Gb FOC to start and if anyone you invite to your dropbox also decides to setup a dropbox account of their own, you get an extra 250Mb...

(Would anyone like a referral ?)

[Zakalwe]

I have uploaded a newer version of the stacked file here. This is a big file (145mb in size) so right-click and "Save as"/"Save link As".

Deep Sky Stacker has an option (under Cosmetics) to detect and fix hot/cold pixels (see here). A by product of this is that it was cutting the cores out of the stars. I have re-stacked, and the stars now look much healthier.

[AndyUK]

The download's not complete yet, but I know it's M101, and I believe you have 200s and 150s subs and 200s darks...(?) Can I just ask how many subs you have of each exposure and at what ISO and also how many darks...? (... and ditto for any flats or bias frames?)

I did have a quick look at the other stack (it downloaded anyway!) and it did seem quite noisy, but I see what you meant about the polo stars

[Zakalwe]

The download's not complete yet, but I know it's M101, and I believe you have 200s and 150s subs and 200s darks...(?) Can I just ask how many subs you have of each exposure and at what ISO and also how many darks...? (... and ditto for any flats or bias frames?)

I did have a quick look at the other stack (it downloaded anyway!) and it did seem quite noisy, but I see what you meant about the polo stars

16 x 200 sec lights, 10 x 150 secs lights and 11 x 200 sec darks. ISO800. No bias or flats. About 1 hour of data.

I need a LOT more data for a dim object like M101 (it was nearly 2AM when I had carted it all back inside last night, so it was late for a "school" night), but I wanted to see what was achievable. I would have gotten more data if I didn't spend so long trying to get a Lifecam to work as a guide-cam.

[AndyUK]

Not surprisingly () I'm really struggling here - After initial levels and curves it seems that the result is fairly mono and there's quite a bit of noise going on and there's a bit of a gradient... Flats would certainly help with that, although gradient xterminator seems to have taken out a lot of it

(Dare I ask, if I send you my email address, is there any chance you can upload all the RAW frames to your dropbox...?)

Your stars are nice and round, so your polar alignment must have been pretty much spot on (), but I'm surprised that with an hours worth of data there's not more available to tease out

Anyway... Here's a very quick first pass, but hopefully one of the "guru's" will be along shortly to wave their magic wand at it and produce something far better...

[Zakalwe]

Thanks for trying Andy....I think that there just isn't enough data in there. More subs needed I think.

[AndyUK]

Hmmm, well you might be right (more subs never hurts!), but I'm sure someone else can get more out of these.

Back in the summer I took 50x90s subs of M31 with a 100mm/f9 - Okay, it is a brighter object, but looking back at those I feel that there really should be more here...

[Zakalwe]

I think that I might need longer subs. Over 3 minutes unguided and I think I am starting to push my luck with the alignment (I used EQMod to PA last night......managed to get that bit sorta right).

[Zakalwe]

This is where I got to with it. Rubbish, I know.

[AndyUK]

Not at all...! As you said, it's a very faint object...

One question though, when saving from DSS, do you save with changes embedded or applied? I know that people use DSS differently, but when I saved with changes applied the result did appear to be much "brighter" than with changes embedded, but, for me, tended to give a "flatter" result in CS... It could just have been me, but after a few months playing about with that option, I've now decided to embed changes and do all the processing work in CS...

[Deneb]

mmm, do you have the complete stack still ?

Nadeem.

[AndyUK]

(Aha... A guru arrives... )

[Zakalwe]

mmm, do you have the complete stack still ?

Nadeem.

Hi Nadeem,

Yes, I have put a link to the stacked TIFF file here:

Clicky

(145Mb file- Right click and Save As)