www.ukastroimaging.co.uk

Questions & Answers/Tips & Tricks => Newcomers Corner => Topic started by: Annie on 12:03:59, 02 December, 2017

Title: NGC7635 image for C & C
Post by: Annie on 12:03:59, 02 December, 2017
Am wanting to improve my basic imaging techniques and also my processing so wanting some constructive criticism on some recent images.  I took this one last week of the Bubble nebula when the moon was around: 21x600sec exp at ISO800 with Takahashi 106ED, crop sensor Canon 550D DSLR and Lacerta MGEN guider.  Stacked in DSS with darks and flats and processed in PS (with my usual limitations!).  The image is uncropped.

Thanks
Annie

Title: Re: NGC7635 image for C & C
Post by: Nimbus on 18:33:01, 02 December, 2017
Hello Annie,

Firstly, you have a pleasing image and it is always less than easy to comment on a Jpg of restricted size,  but here are some observations from me:

You mention you applied Darks and Flats - no Bias? Bias are need also.

There appears to be a dark circle at about 7 o'clock to the bubble  indicating the Flats may not have dealt with any artifacts.

Overall the image appears a little dark for my taste indicating there may be some black clipping.  Try aiming for a background of about 30-40 in the info palette.

I suspect there is more nebulosity hidden in the original image that means you can stretch things a little further.  You may wish to remove the stars first to prevent bloating.  You can layer the stars back in and tighten them up to reduce there dominance in the image.

The colour balance appears a little off, try aligning the left hand side of each of the RGB histograms to achieve the correct balance.

The good news is you have retained some colour in the stars which can easily be lost when stretching.

Have a go at adjusting your original image in line with the above and re-post and we can see where we go from there. If you are not sure how to achieve the above any of the techniques - ask away.

HTH Jim
Title: Re: NGC7635 image for C & C
Post by: Annie on 19:24:24, 02 December, 2017
Hi Jim - thanks for some helpful comments.  Firstly I just forgot to mention the bias but I did add them and you're right that the flats didn't get rid of all the artefacts.  I take my flats using a head torch with something over it to diffuse the light but my torch packed up so the flats didn't work very well! I can try most of what you suggest other than removing and layering the stars back in but I guess that would be a bit difficult to explain so am happy to try and work that out!!

Thanks again Jim

Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 14:31:12, 06 February, 2018
Another version of my original 'Bubble'.  First go with any image at extracting stars - via Straton software - to enable further stretching and later trying to dim and reduce the impact of the stars by using the minimum filter method in photoshop a la Adam Block tutorial.  I obviously need loads of practice as though I may have cropped too much, some of my stars, esp bottom right seem mis-shapen and the image lacks sharpness.  Before I try again can anyone see an obvious cause of this?

Thanks
Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 14:31:51, 06 February, 2018
Oh dear wrong image!

Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 14:38:41, 06 February, 2018
Hope this is the new version!

Annie
Title: Re: NGC7635 image for C & C
Post by: Carole on 20:59:49, 06 February, 2018
Hi Annie, I am rather dubious about the head torch method of flats,  has it always worked for you previously?

Anyway, is there any chance of posting up the unprocessed final tiff stack so we can have a "go" at processing it and see what there is in the image, then we can tell you how we did it.  I think this would be the most useful thing to do.  You might have to link it via dropbox or something.

Big improvement between the first and 2nd processing posted up.

I thought you were getting a mono camera, or am I wrong?

Carole
Title: Re: NGC7635 image for C & C
Post by: Annie on 08:00:56, 07 February, 2018
Thanks Carole - I'll send a link for dropbox.  I'm not sure about the head torch method either so last imaging session had a go with a light panel which I think will be much better.

Yes you're right I was thinking about a mono camera but my processing skills were so lacking I have been concentrating on improving those.  Inspite of  all the advice towards a mono ccd am still of a mind to buy a OSC camera as a first step which won't preclude me from moving to a mono ccd( when the weather and my skills improve!) - was thinking about an Atik 460ex, colour in the first instance.

It will take a bit of time to upload the image to dropbox - we have a very slow broadband speed.

Cheers
Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 09:47:38, 07 February, 2018
Having problems uploading the file into dropbox. Not had this problem before - after about half an hour it says an error has occurred during upload!  No idea what's causing this at the moment.  Am not exceeding my dropbox quota so wonder if it's my slow broadband speed so will try later in the day!
Title: Re: NGC7635 image for C & C
Post by: Annie on 16:02:17, 07 February, 2018
It's unlikely that I'll be able to provide a link to dropbox at the moment as I suspect my miserable upload speed of 0.2mbps is why the upload keeps failing!  How frustrating is that!!!! 

Annie
Title: Re: NGC7635 image for C & C
Post by: Carole on 09:36:46, 08 February, 2018
That's a shame Annie, well tell us what your usual processing routine is and we can feedback C&C.

Carole
Title: Re: NGC7635 image for C & C
Post by: Annie on 10:06:54, 08 February, 2018
Thanks Carole - have only recently made a 'real' effort with processing so am in the early stages of establishing a 'workflow'.  The bubble was my first attempt to apply an initial stretch using levels and curves and correcting the colour balance  by aligning the histogram through use of the channels in levels ( Steve Richards suggests this in his new book 'Dark Art or Magic Bullet', mentioned in Nimbus' useful books thread).  I then used Straton software to remove the stars , converted the three greyscale images produced by Straton to a colour RGB, did a further stretch of the starless image before layering the original back onto the 'stretched starless'!

However I've been looking at my original light frames and to my eyes the stars look perfectly ok but not when I look at the stacked tif produced by DSS.  Am thinking I should re-stack but without flats and bias?

Have attached JPEG versions of an individual light frame and the stacked version of 21x600sec exps.  What do you think?

Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 10:08:13, 08 February, 2018
This one is the stacked image.

Annie
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 10:42:48, 08 February, 2018
Annie

Stars don't look too bad in either. As mentioned in another thread the Tak106 can be a bit temperature sensitive on focus (the much older ones were less so) so focus can wander as the temperature drops. Not at all unusual for the stars in the stack to bloat a little as a result. Also not unusual to loose apparent star colour in the star. Its there, it just needs a good oomph of saturation.

DSS is getting a little old in the tooth now and things have moved on since it was last updated. AstroPixelProcessor is well worth a try. It works well using the default settings on images such as this and some of its trickery may help keep stars under control a little better. Well worth downloading the trial version and giving it a go. I am using it more and more for my stacks.

Chris
Title: Re: NGC7635 image for C & C
Post by: Annie on 12:48:05, 08 February, 2018
Another in the long list of 'keep meaning to' is to consider another stacking process so will have a look at the one you suggest Chris.  Have been working through some tutorials of Adam Block - 'Cosmic Canvas' and he uses something called 'CCD stack' - looks very sophisticated to me. 

Annie
Title: Re: NGC7635 image for C & C
Post by: clive on 13:32:03, 08 February, 2018
Have a look at astroart  ive been useing it for a while  its very easy to use  and its very fast at stacking
 It cost 149 pound
First light optics
 
Clive
Title: Re: NGC7635 image for C & C
Post by: Annie on 14:34:12, 08 February, 2018
Thanks Clive - I think Carloe uses Astroart and says the same so will definitely look at that.  It's the phrase 'very easy to use'  I like!!

Annie
Title: Re: NGC7635 image for C & C
Post by: Carole on 15:04:21, 08 February, 2018
Yes I use Astroart, its very fast.

I agree with the approaches in Levels and curves as described in your recent posts (recommended by Steve Richards) that's what i always start with.

I saw you mention Adam Block.  When I first started I bought a DVD of his on post-processing.  Making every Pixel count.
Much of it went way over my head to start with, but as time went by I got to grips with many of the amazing things he can do to bring out the best in an image.  I don't even need to refer to it any more, I could probably put it up for sale.

I had a little stretch and colour balance of the image you had started to stretch (the one that appears a little blue).  It's difficult working with a Jpeg, but I can see there is a fair amount of data there ready to be pulled out.


Title: Re: NGC7635 image for C & C
Post by: Annie on 15:29:44, 08 February, 2018
Yes Carole there is a lot to be brought out - am going to keep practising on it.  I bought Adam Block's 'Cosmic Canvas' . Mine came on a memory stick.  A fair bit is about CCD imaging but a lot of techniques can be applied to RGB and it's all done in photoshop apart from his stacking software which is a sophisiticated looking 'CCD Stack'.  The more I look at his tutorials  the more I tease out and he really does start with a few very basic essentials before moving onto more advanced stuff - something for everyone level.  Well constructed, logical progression, comprehensive and very clear presentation. It was Nimbus who first told me about it.  Not cheap but still good value I think.

Thanks for the processing and comments Carole.

Annie
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 17:08:04, 08 February, 2018
...Have been working through some tutorials of Adam Block - 'Cosmic Canvas' and he uses something called 'CCD stack' - looks very sophisticated to me...

It is and to my mind overly complicated.

Chris

ps Adam in now a PixInsight convert :-)
Title: Re: NGC7635 image for C & C
Post by: Annie on 17:26:09, 08 February, 2018
It looked very complicated to me Chris  but then a lot of things to do with astrophotography look that way to me - well at least initially.

4 hours to upload a 162mb file  but here it is: https://www.dropbox.com/sh/ag9q07ge21l5qop/AAB_DLeTZQ3_z2VOTpSNsHk_a?dl=0 (https://www.dropbox.com/sh/ag9q07ge21l5qop/AAB_DLeTZQ3_z2VOTpSNsHk_a?dl=0)

Annie
Title: Re: NGC7635 image for C & C
Post by: Nimbus on 17:32:18, 08 February, 2018
At the risk of causing an international incident or something even more important and to restore balance Adam uses both PS & PI. Handy when you have access to a mountain observatory and lots of expensive gear including software which come as part of ones job. But then I do love his work!

CS Jim
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 19:26:56, 08 February, 2018
Annie

For a DSLR that is a nice widefield. Something a bit odd with the stars in the bottom right hand (lots of doubles/triples) so either you have some errant subs or DSS really is not working well.

Given a bit of love and care I got this out of it.

Chris
Title: Re: NGC7635 image for C & C
Post by: Annie on 21:03:57, 08 February, 2018
Brilliant Chris - now I know what to aim for! Taken with modified Canon 550D that I had given to me and had to glue the grip back on! (And the Tak 106ed of course). I've had a look at some of my subs but will go back and check through them all now.  I have in any case been meaning to move to another stacker so now would seem a good time.

Thanks again Chris.

Annie
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 09:02:20, 09 February, 2018
Annie

Looking at the stack again I do wonder if something is shifting in your image train. DSS is able to stack stars in the top left of the image but has increasing problems towards the bottom right. That feels to me as though the camera has tilted with respect to the rest of the train during the capture run. any chance of posting up a picture of your image train from focuser to camera. My money would be on the T-Ring. They often come in two parts with the tiniest grub screws know to humankind holding them together. If they are not nipped up tight, the two parts can move relative to each other.

Justa thought.

Worth checking through the subs and looking at the bottom right of each to see if this happens at all mid capture. Would be really helpful to see all the subs in this image but I realise that might be difficult with a slow internet connection...mine isn't much better!

Chris

ps stacking with software that includes distortion correction may well bring those errant stars into line whereas DSS will just try and match as many stars as possible as is.
Title: Re: NGC7635 image for C & C
Post by: Annie on 11:39:57, 09 February, 2018
Chris - have looked at all my subs and can see no sign of the the stars being anything other than fine - there seems to be little if any difference in their shape between any of them which would seem to suggest that there was no slippage in the imaging train though am happy to post a photo of it?  The change appears after stacking?  I think I'll go back and stack again with calibration files and perhaps once without flats and bias. If you think it will be helpful am happy to convert my subs to JPEGs and see if dropbox will upload them?

Am now minded to look more closely at some of the imaging I've done since the Bubble nebula!!! Am also going to look at Astroart and Astropixelprocessor and make a commitment to one of them.

Thanks for your help.
Annie
Title: Re: NGC7635 image for C & C
Post by: Carole on 11:45:19, 09 February, 2018
OK, stage 1, I processed your image using the following processes in Photoshop.
(N.B. Using a mono camera you could get far more detail out of it, but that's for another day).

1) Image adjust/levels, bring the back and grey sliders in to the histogram but not touching it.
2) Crop black edges
3) 2nd iteration of levels using grey and black sliders.
4) balance the colour
5) Change to 16bit (whoops missed that).
6) Curves, bring in black slider to histogram.
7) Curves (anchor B & W points and lift nebulosity
8) Image adj/match colour - increase colour intensity
9) Further stretch in curves again anchoring B & W points.

I will comment on the next post when I have uploaded the images to Astrobin regarding the issue with the stars (which Chris also found).

Annie, a comment on Astroart for stacking, I have only ever used it for stacking mono images, never had any success with stacking colour data, and I just googled it and it looks very convoluted for colour. 

Link to Astrobin to see larger version
https://cdn.astrobin.com/thumbs/gbTdueFRCeiX_1824x0_wmhqkGbg.jpg

(https://cdn.astrobin.com/thumbs/gbTdueFRCeiX_620x0_wmhqkGbg.jpg)
Title: Re: NGC7635 image for C & C
Post by: Carole on 11:59:00, 09 February, 2018
Quote
have looked at all my subs and can see no sign of the the stars being anything other than fine - there seems to be little if any difference in their shape between any of them which would seem to suggest that there was no slippage in the imaging train though am happy to post a photo of it?  The change appears after stacking?

That is bizarre.  I am not entirely convinced it is a tilt error as you would expect the stars to be elongated to one corner.

In the next image I have highlighted all the areas where the stars are mis-shapen and bizarrely the bottom right hand corner where you would expect it to be the worst - the stars are round.  Also I have never seen the misshapen stars extend accross the centre of the image with tilt.  Additionally the stars are not just elongated but they have a pair of stars either side of them.  If this was a dodgy  sub in the pile (moved through wind or nudging or something), it would be all over the image).

Something else is going on here and I am inclined to go along with your point that your original subs are OK but the problem happens after stacking.  I have seen some weird results previously with DSS.  I don't know what to suggest as an alternative for stacking, but maybe its time for the purchase of your Atik460EX so you can leave these troubles behind.

In the meantime if you can get the subs to Chris so he can try it in Pixinsight it would be a good step as we really need to find out what is really going on here. 

ignore plate solving lines on Astrobin.  As you can see the worst stars are not just confined to one Corner.

(https://cdn.astrobin.com/solutions/26c0c5e1-dd5e-4bef-bb92-a08d352dcb05-1518179151.png)

Here is a close up of the stars over the Bubble.  (Ignore the plate solving lines)
(https://cdn.astrobin.com/solutions/d75db810-ac5a-42e0-914c-4983384d10ff-1518177663.png)



Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 12:07:51, 09 February, 2018
Annie

I dont think its anything to do with flats or bias, it is something of a physical nature. Most of the image stars are perfect but the bottom left corner is distorted. That is quite strange indeed.

How many image files (RAW) are there to upload to Dropbox? Pixinsight has tools to analyse such things but I'm a bit lost without the data.

Chris
Title: Re: NGC7635 image for C & C
Post by: Annie on 12:18:34, 09 February, 2018
21 raw files Chris.  I can't think that it's anything to do with calibration files either.

Am happy to have a go at uploading the raw subs - it would be re-assuring to find out the cause.

Annie
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 12:19:05, 09 February, 2018
Annie

Analysing the stack the eccentricity does increase in the areas where the multiple stars happen and bizarrely the field flatness is worst in the middle rather than towards the edge. Your 106 is a Petzval design and the flatness does change depending on focus, I see it on my 85 but I tend to loose flatness in the corners as focus wanders. It COULD just be done to focus wandering but I'm not convinced.

I really am impressed with Astro Pixel Processor, getting a hold of the trial and just loading in the light frames and running it all at default values would be interesting.

Chris
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 12:20:54, 09 February, 2018

Am happy to have a go at uploading the raw subs - it would be re-assuring to find out the cause.


I think that would be a good idea even if one at a time. I can run them through PI and APP and see what happens then. PI in particular has tools that will report on stars that are in positions away from the mean so that would be the first port of call.

Chris
Title: Re: NGC7635 image for C & C
Post by: Carole on 12:31:55, 09 February, 2018
Chris has been busy while I have been busy stacking, processing and analysing your images Annie. 

I will leave it to Chris now as I don't have the tools that he has to hand to work out what is going on.  Hope you get to the bottom of it and will pop back in to the post to see if the cause has been found.

Hope the processing steps I anotated when I processed your image will have helped with your processing.

Carole
Title: Re: NGC7635 image for C & C
Post by: Annie on 12:46:27, 09 February, 2018
Yes Carole , Chris has been busy! 

I appreciate the processing notes and am sure they will be helpful.  If I don't see you 'popping in' I'll let you know the outcome.

Many thanks
Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 12:49:29, 09 February, 2018
Chris - I'll start uploading - don't hold your breath!! And also download the trial version of astropp - don't hold your breath with this either - am certain to take a while to get to grips with it, how long's the trial version last!!!

Annie
Title: Re: NGC7635 image for C & C
Post by: chris.bailey on 13:59:49, 09 February, 2018
Annie

Its a 30 day trial but the author Mabula is pretty friendly to extend that if you ask nicely!

If I get five minutes I'll post some screen shots of how to load everything up and the settings I have used. It really is pretty simple once you get used to it and you can ignore most of the mumbo jumbo.

Chris
Title: Re: NGC7635 image for C & C
Post by: Annie on 15:57:01, 09 February, 2018
Cheers Chris!

Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 09:06:34, 11 February, 2018
It's proving impossible to upload to anything to Dropbox given my broadband download/upload speeds are so slow so have given up!  Have since imaged the Horsehead nebula with the same kit and stacked with darks/flats/bias and as everything looks fine I'll consider what happened with the Bubble nebula as an anomoly and worry if it happens again!  I am however going to download a trial version of Astropp and also have a look at Astroart.

Thanks for all the help, I've learnt a lot.

Annie
Title: Re: NGC7635 image for C & C
Post by: Annie on 19:37:57, 22 February, 2018
I went back to basics and restacked this image and much to my surprise got rid of the weird stars in the bottom right hand corner by simply stacking with darks only!!  Have since imaged another target and stacked with flats and bias with no odd results. 

Anyhow this is as far as I can go with the image at my skill level but it's undoubtedly a vast improvement on the image I first posted asking for help with processing so a big thank you to everyone who took time to repond -much appreciated.


Annie
Title: Re: NGC7635 image for C & C
Post by: Carole on 16:03:16, 23 February, 2018
Big improvement Annie, well done.

Carole
Title: Re: NGC7635 image for C & C
Post by: Annie on 16:23:26, 23 February, 2018
Thanks Carole  and for all your help!

Annie
TinyPortal © 2005-2012