Ha was taken on the first night when there was 8mps wind.
6 x 1200secs Atik383L Ha 12nm ED80 on NEQ6 captured in Artemis cooled to -10
Another huge object I have not imaged before.
(http://www.caroleastronomy.webspace.virginmedia.com/_wp_generated/wp64f67389_0f.jpg)
The DSLR image was then taken over 2 nights with reduced and then no wind on the 2nd night.
total of 37 x 5mins Modified Canon 450D ED80 captured in APT with dithering.
(http://www.caroleastronomy.webspace.virginmedia.com/_wp_generated/wp3d5deed1_0f.jpg)
The combined Ha and DSLR image did not come out too well, due to the alignment, and not happy with the colour, so I won't be posting that.
What a huge difference in detail between the DSLR image and CCD image.
Carole
Again, a very interesting comparison. There's so much detail and texture in the narrowband image.
But I wonder how the Canon image would look if you displayed the red channel in monochrome?
In any case, its an excellent California Nebula!
Mark
Thanks Mark.
QuoteI wonder how the Canon image would look if you displayed the red channel in monochrome?
This is the Red channel in Mono:
(http://www.caroleastronomy.webspace.virginmedia.com/_wp_generated/wp87887b4c_0f.jpg)
Following some new processing techniques I learnt at IOW this weekend I have now combined the Ha and DSLR data. As you can see from the other images I did at the time they came out very pink and I have now managed to produce a combined colour closer to the DSLR image as opposed to the horrible pale pink colour I was getting before. I have had to crop this to align the images.
(http://www.caroleastronomy.webspace.virginmedia.com/_wp_generated/wpa88440c8_0f.jpg)
That's come out well Carole, though perhaps a bit saturated.
There's also an article in November's Astronomy Now by Nik Szymanek about mixing broadband and narrowband data, but I haven't had a chance to read it yet.
Mark
It's a bit better Carole but the colour data in the image is still wrong. All of your stars are pink. There are no blue or orange stars. The two stars top left of the dark 'hole' in the nebula should be both blue and orange as should the two bottom right of it. In your image they are identical colours.
Take a look at this image to see how the star colours should be...
(http://astro.nightsky.at/Photo/Neb/NGC1499_WN.jpg)
Hi Mark, thanks, I'll try to get hold of a copy of AN. Yes I was wondering whether it was too saturated myself but I had been looking at it for too long, so decided to see what every-one else thought.
Hi Mike: Thanks again, yes I struggle with star colours even with just plain DSLR, though I don't see the stars as pink in my image, I see them as white. Ha as you know captures in mono and DSLR obviously in colour which I struggle with as regards star colour (see the DSLR image above - the stars are mainly white), but I am working on this and might re-visit CN and see if I can improve it at all. Though majority of the stars in this image are from the Ha image and only just found out how to grab the stars back from the RGB image.
Big new learning curve is combining filters and RGB.
Carole
I think you should concentrate on getting the colours right with the DSLR image only first then you can work on adding the Ha as the luminance layer. If the colours in the DSLR image are not right before you add the layer it will not come out right.
Maybe you can put your raw files up somewhere so we can download them and have a play?
Quote from: Mike
Maybe you can put your raw files up somewhere so we can download them and have a play?
The raw stack (linear scaled) would be sufficient i.e. without logarithmic or histogram stretches or any other kind of "helpful" scaling applied. A crop would do.
QuoteMaybe you can put your raw files up somewhere so we can download them and have a play?
That would be good Mike, I'll sort something out, though it won't help me unless Photoshop is used as I don't have Pixinsight and Iris is too difficult to understand.
Also finding problems with the differing stars sizes between the Ha and DSLR images.
Not sure what you mean by linear scaled Mark.
Carole
What I mean is that if you are using DSS for stacking, for example, then we want the stacked TIF without DSS doing any "funky stuff" before you save it.
Iris really isn't as complex as it "looks" unless you start really messing with is.
The colour stretch tool will do wonders for the star colours and is very simple. Mark explained it too me at DSC in under 15mins!
Yes I've tried the colour stretch tool in Iris John, and a couple of other things, it seems to want me to input commands and statistics and I don't have a clue what to put in.
Carole
QuoteWhat I mean is that if you are using DSS for stacking, for example, then we want the stacked TIF without DSS doing any "funky stuff" before you save it.
I don't normally do any post-processing in DSS except slide the colour channels together so they are on top of each other. If you don't want that I'll have to do a re-stack and of course the Ha won't have had that done anyway.
Carole
Quote from: Carole
I don't normally do any post-processing in DSS except slide the colour channels together so they are on top of each other.
I'm not familiar with it (I don't use DSS) but that's probably OK.
I think a processing session to 'pool' knowlage should be arranged for some time in the near future....
*Hijack alert*
That would be a hefty session and who would do the pooling
Folks have been using Registax, DSS, Iris, PI and PS to get good results
That's a good point, although I didn't mean quite as general as that...
More like the imaging session at Roberts when he spoke about pixinsight.
I hope I have the right files here:
this is the DSLR stacked file:
https://dl.dropbox.com/u/5730788/California%20Friday%20DSLR%2024%20x%205mins%20800%20ISO%20ED80.TIF
and this is the Ha stacked file:
https://dl.dropbox.com/u/5730788/California%20Thursday%20Atik%20Ha%206%20x%201200%20-10%20degrees.TIF
Both stacked but unprocessed you'll have to flip one of them over and align them.
Carole
QuoteI hope I have the right files here:
Nope,
the top one is a 404.
i.e. no file.
Bottom one is ok
Scratch that, the top one was probably still uploading, as it is 70Meg!!
Both are working now.
Oh the joys of rural broadband.
97 mins and counting. :roll:
OMG!
Quote from: Mac on Oct 22, 2012, 15:01:14
Oh the joys of rural broadband.
97 mins and counting. :roll:
2.5min from my rural location ...
The colour data from the DSLR data is a nightmare! It just doesn't "feel right" to me at all - it doesn't respond in the usual way. I can fully understand why you couldn't get star colours out of it.
I really think that something strange has happened somewhere in DSS. Before making a final judgement, I'd really like to see the same data stacked in IRIS - we must arrange this somehow.
Mark
I would agree with Mark.
The background has come up much redder than I would expect plus the is a lot of horizontal banding. It's also a lot noisier than I would have expected from that many subs.
Just a few thoughts - Can you check if there are any rogue subs i.e. those where any clouds may have crossed or short subs included accidentally. It's important that all the subs are as identical in quality as possible. Were the darks done at the same temperature ?
Robert
i would like a lesson in IRIS..................
Mark, i was having a practice and did not see how I could include my flat darks. is there a way?
I also also couldn't resist but have a play. Sorry, but i'm getting a little tired of looking at my M31!
As was said before, i found horizontal banding in the DSLR image and it seemed very noisy, although i did manage to eventually eek out the colour and calm the noise etc, possibly at a loss to fainter detail but i would still call it a respectable result.
As with the Ha one, it just did my head in! I could hardly get anything useful out of it and had i nightmare getting colour into it! But i possibly gave up to soon and it was my first ever play with an image from a monochrome camera...
As with combining them i would't even know where to start!
Like mark said, would be really good to have a look at a stack from Iris, or have the individual subs to play with as then i'd have a good idea on how to combine them and would like to see how they would work with some of the HDR stuff i have been messing with.
Perhaps at the meeting this Thursday (it is Thursday right?) You could bring it along on a memory stick any anyone interested in fiddling should bring a laptop and take a copy of the subs...
Quote from: Fay
i would like a lesson in IRIS..................
Mark, i was having a practice and did not see how I could include my flat darks. is there a way?
Yes - try this tutorial:
http://www.saratogaskies.com/articles/cookbook/index.html
It's where I originally learnt to use Iris!
Carole,
Quote from: MarkS
Quote
I don't normally do any post-processing in DSS except slide the colour channels together so they are on top of each other.
I'm not familiar with it (I don't use DSS) but that's probably OK.
Unfortunately I was wrong!
I woke up in the night and realised what is going on in your image. Those RGB sliders in DSS are
very dangerous for the integrity of your data. Don't touch them with a bargepole! They apply a completely non-linear RGB transformation to the data and it is the
wrong way to adjust for the RGB balance of your camera. Sorry, I don't know DSS well enough to tell you the correct way to adjust colour balance.
Right click on this test image and save it to your machine
http://www.markshelley.co.uk/webdisk/Ramp.tif (http://www.markshelley.co.uk/webdisk/Ramp.tif)
It contains more than 50 shades of grey. :cheesy:
There are no shades of any other colour.
Open it in DSS and adjust the RGB slider bars then resave it as something else. Now open it in Photoshop or anything else where you read out the pixel values. You'll notice that evey pixel now has a different colour i.e. a different ratio of R,G and B. i.e. DSS has happily gone in there and introduced a whole range of colours that were not there before. Conversely, when you adjusted the histograms so they were on top of one another, you unknowingly destroyed star colour that was previously there.
Try restacking in DSS and save your image
without playing with slidey bars or anything else. Put the result in dropbox and I'm sure we'll have sensible data we can play with!
This is just one example of the huge number of ways available to destroy the colour in your stars. This is why I also bang on incessently about using the Iris asinh colour stretching to perform data stretch. The Iris asinh function is the only function I know that stretches the data values whilst preserving the original ratios of R:G:B within each pixel. If your stretch function is not preserving those original ratios then it is probably destroying valuable colour data.
If it's any consolation, PixInsight also does a excellent job of destroying star colour. It's probably my lack of experience but I found that all the obvious ways of applying some kind of data stretch also changed the R:G:B ratios in the pixels :-(
Mark
This is all very useful Mark, I have been wondering for some time whether DSS is doing the job properly, particularly as it sometimes doesn't apply flats properly, I have been wanting to find a different stacking programme for some time, but because I found both Maxim and Iris too complex and didn't have a paid copy of Nebulosity (and didn't want to fork out until I was sure I would be able to use it), I have been a bit stuck. I have now got a tutorial for Maxim and am starting to learn this. I will also take a look at Iris again now you have supplied a link and JonH has also sent me some useful info.
It seems daft to me to provide a tool in DSS (the colour adjustment tool) which has the result of destroying data and would explain why I have had a lot of problems with star colour.
I will re-check all the subs for rogue ones, though I do normally check at the time of taking them (there is only a couple of hours subs in this stack by the way as I was just going to use it for the colour component of this image and not as a stand alone result.
So looks like I have been fighting a losing battle for quite a long time.
I am working today so will take a look at all this tonight. Thanks for doing all this guys, and I think this is not only a very useful exercise for me, but might be something we could do on a regular basis, I am sure people like myself and new imagers would find using other people's data, particularly that of more experienced imagers a really useful exercise.
Thanks
Carole
Quote from: MarkS on Oct 23, 2012, 06:10:30
If it's any consolation, PixInsight also does a excellent job of destroying star colour. It's probably my lack of experience but I found that all the obvious ways of applying some kind of data stretch also changed the R:G:B ratios in the pixels :-(
That certainly needs investigating have you any examples ? Of course you could always apply asinh stretching within PI if needed but I take it you were using some of the standard tools. I'll start a PI thread under technical...
Quote from: RobertM
That certainly needs investigating have you any examples ?
I'll do better than that - I'll create a 16 bit synthetic example TIF tonight - I need some test data in any case. Besides, it's the only way of working out what on earth various processing functions in Iris, DSS, PixInsight, Maxim and (not forgetting) Photoshop are doing.
The image will contain a row of reddish stars (all with precisely the same R:G:B ratio) with different magnitudes from the dimmest up to completely oversaturated - blurred with a Gaussian to simulate atmospheric effects. Then a similar row of white stars and a similar row of bluish stars. The whole thing will be given a synthetic RGB colour balance as if taken by a typical DSLR sensor. It will also have a uniform background sky glow and a hidden faint picture that needs to be revealed.
The goal of the exercise will be to use standard tools in each application to correct the RGB balance, remove the background and then stretch the image to reveal the hidden faint picture whilst maintaining correctly coloured stars. We then document the required processing sequence for each application.
Iris will win this hands down; in Photoshop I bet it's not even possible; the rest will fall somewhere in the middle.
Quote from: MarkS on Oct 23, 2012, 11:26:34
Quote from: RobertM
That certainly needs investigating have you any examples ?
I'll do better than that - I'll create a 16 bit synthetic example TIF tonight - I need some test data in any case. Besides, it's the only way of working out what on earth various processing functions in Iris, DSS, PixInsight, Maxim and (not forgetting) Photoshop are doing.
The image will contain a row of reddish stars (all with precisely the same R:G:B ratio) with different magnitudes from the dimmest up to completely oversaturated - blurred with a Gaussian to simulate atmospheric effects. Then a similar row of white stars and a similar row of bluish stars. The whole thing will be given a synthetic RGB colour balance as if taken by a typical DSLR sensor. It will also have a uniform background sky glow and a hidden faint picture that needs to be revealed.
The goal of the exercise will be to use standard tools in each application to correct the RGB balance, remove the background and then stretch the image to reveal the hidden faint picture whilst maintaining correctly coloured stars. We then document the required processing sequence for each application.
Iris will win this hands down; in Photoshop I bet it's not even possible; the rest will fall somewhere in the middle.
I really like this idea!
Just thought I'd add my two cents worth, with DSS I think it might also be a good idea to uncheck the align RGB option in the stacking settings.
Although I found having this checked made it a little easier and quicker to process the image I also found I could get better results if I did it myself outside of DSS.
Perhaps this is just me but maybe this is also manipulating the image un an undesired way?...
Sounds like a challenge.
Im up for the photoshop bit.
Mac.
Quote from: Mac
Im up for the photoshop bit.
I'm very happy for you to prove me wrong regarding my negative comments about PS!
Perhaps an Iris and/or PixInsight workshop at one of the Imaging Sessions or perhaps we should actively set a date for another Imaging Workshop?
well i managed to get this from the original tiff, (not including the HA).
Photoshop only.
(http://farm9.staticflickr.com/8046/8116403479_b5f2829c85_c.jpg)
Cropped and scaled, needs a lot more work, but this was only about 5 mins of playing.
and i was just seeing what data was there.
Mac.
Well this is what i got out of just the DSLR image, I spent quite a bit of time trying to get this right and worked it in both Iris and Photoshop.
I think the colour balance is still a bit out but the star colours are starting to show threw and think i managed to get the worst of the noise under control.
That banding refused to go away entirely though and can still see bits of it in places!
(http://farm9.staticflickr.com/8044/8116646822_8f25103f6a_c.jpg)
Well this just goes to show Carole that overall you are capturing good data, it is just a case of processing im such a way that it comes out looking good. capturing the raw data is the easy part, it is the processing that is the 'magic arts'.
Quote from: Mike on Oct 23, 2012, 19:45:06
Well this just goes to show Carole that overall you are capturing good data, it is just a case of processing im such a way that it comes out looking good. capturing the raw data is the easy part, it is the processing that is the 'magic arts'.
I would say the data was ok, it was a pig to work with though and didn't respond very well to processing. There is defiantly something not right with it, I would normally apply a MUCH harsher asinh stretch in Iris to really bring out the star colours but this just didn't like it at all.
I would REALLY like to have another pop at this but stacked in Iris, what Mark has said kinda makes sense to me so a comparison should revel a huge differance!
Thanks for doing that Mac and Jon, I will now do a newly stacked image without using the RGB controls in DSS and I will also post up the individual subs so you can have a "go" in your favourite stacking software.
Mark I would also be interested to see what you can get from it with Iris, plus has any-one had a "go" at adding the Ha data yet to the DSLR data?
Carole
Here's my attempt in PI (DSLR only). I would be tempted to mask the stars and colour enhance them more separately but for this exercise I've only used whole image processing i.e. not processed any part individually.
(http://farm9.staticflickr.com/8332/8116898225_cd12369576_b.jpg)
Full monte: http://farm9.staticflickr.com/8332/8116898225_c5befc3fbf_o.jpg (http://farm9.staticflickr.com/8332/8116898225_c5befc3fbf_o.jpg)
Process flow for this was:
Canon banding reduction (default options)
Background neutralisation (default options)
Colour calibration (default options)
Multiscale Median Transform (a little tweeking)
Histogram Transformation (defaults copied from 'Screen transfer function' and all colour channels equally stretched)
Colour saturation (a little tweeking)
No masks or any other special processing.
Robert
Robert, please tell me thats not just the DSLR data! :o
That's really good Robert, just goes to show what can be done in expert hands.
Carole
Thanks Carole but in expert hands it would be much better. Do you want me to try to add the Ha Data ? I've never done that in PI before so should be good for a laugh.
Robert
Yes please Robert.
carole
OK I have checked the subs and re-stacked.
I have done two slightly different versions, but neither of them have had any post DSS adjustments to the RGB.
Version A. I used the recommended settings in DSS as follows:
AHD debayering
Reset all white balance settings
BIAS set black point to 0
Masterdark - Sigma clipping median combination
Bias - Median combination
If background colour is hard to fix tick background calibration (I chose this in the A version).
The RGB sliders were on top of each other at the end.
Version A link:
https://dl.dropbox.com/u/5730788/Califronia%20stack%20A%20RGB%20ticked.TIF
Version B: I used the same settings as above but tried bayer linear interpolation instead of AHD which I normally use.
I did not tick the Background colour is hard to fix option and the RGB sliders were separate at the end.
Version B link:
https://dl.dropbox.com/u/5730788/California%20stack%20B%20RGB%20NOT%20ticked.TIF
Increasing the saturation early on in the processing is producing better star colour. I think I left this too late previously.
If any-one would like the actual RAW files, I will have to link those to an E mail address, let me know if you would like this as I'll have to PM it to you as you can't link to a folder of files in Dropbox in the Public folder. If I don;t have your E mail address please PM it to me.
Carole
Version A "felt" the easier to process but I didn't have time to perform a quantitative assessment. I'd like to do so, to understand precisely what are the differences between all 3 versions.
There seems to be something odd with the stars at the top right in all three versions - they have a reddish fringing. Was this the corner of the original image - possibly with a slight optical aberration?
Not quite sure which stars you are referring to Mark, there is a very slight dither overlap, but only right at the edge. Again will examine closer when I get home from work today.
Meanwhile this is my re-stack and process of Version A I didn't really get much difference between A & B, and both of them are suffering with horizontal banding (tried Noel's actions horizontal banding redution 3 iterations but it's made no difference), and this is one of the reasons I decided to get a CCD camera as this is becoming a real problem - that and noise.
I have managed better star colour by increasing the saturation at an early stage.
(http://www.caroleastronomy.webspace.virginmedia.com/_wp_generated/wp1a3d1e0a_0f.jpg)
Star colours are much better! It's a big improvement.
Thanks Mark, I think I've worked out how to keep the star colour now, do an initial levels followed by an increase in saturation before going on to any more stretching.
Carole
I've taken all 3 of the TIFF files you have posted, subtracted one from another, divided one into another and have come to the conclusion that Version B is the least adulterated.
In your original image, you had manually scaled the RGB before the background had been subtracted, I guess it's using the black and white points selected by DSS. This scales the data in a non-linear manner which I don't think is ideal.
Version A has done an RGB auto scale and come up with something different to what you did manually in the original image. It seems to have scaled RGB in such a way as to make the background uniform grey whilst preserving the white point. This is preferable to the original.
Version B is my favourite because it seems to have completely left the stacked data unadultered (though I'd like to verify this against one of the original subs). This allows you to make all your processing choices post-stack. Definitely my preferred way of doing things!
Mark
Many thanks for all your effort and analysis Mark, that was certainly worthwhile.
Does any-one have any ideas how to deal with horizontal banding in Photoshop?
Quote(though I'd like to verify this against one of the original subs).
Mark would you like me to post up one of the RAW subs in dropbox, or indeed do you fancy stacking the whole lot in Iris?
Thanks
Carole
Quote from: Carole
Mark would you like me to post up one of the RAW subs in dropbox, or indeed do you fancy stacking the whole lot in Iris?
I'll bring a USB memory stick to the meeting tonight. Do you have the flats and darks and maybe the bias frames also?
Hi Mark,
Unfortunately I can't come to the meeting this evening, I have to pick up a friend from Gatwick airport. I'll drop the subs in dropbox for you and send you the link. I'll include Lights, darks, flats and bias.
Carole
That's a better result for star colour but the nebula now has a really strange colour hue.
Robert
I'm more concerned about the awful sky background and horizontal banding Robert. However I now have my LRGB filters for the CCD camera so I think I will put my energy into getting used to using those for the time being when the skies allow.
Should I be using an additional LP filter as well as the LRGB when imaging from Bromley. I asked Bern and he said "see how it goes".
Carole
Quote from: Carole
Should I be using an additional LP filter as well as the LRGB when imaging from Bromley.
In my opinion, stick to narrowband from Bromley.
You're probably right Mark, but I would like to at least see what they will produce from Bromley, even if it's just to practice using and combining them and to get some comparison to using them at a dark site.
Carole
Carole, please can you send me the files of California?
Thanks
OK Fay thanks, also sorry Mark, not got around to doing it yet.
I'll do it tomorrow.
Carole
Processed with PS
(http://farm9.staticflickr.com/8335/8135643038_24a1fd8da0_o.jpg)
Here's my version using your lights, darks and flats:
(http://www.markshelley.co.uk/Astronomy/2012/carole_california_small.jpg)
I only used the DSLR data - not the H-alpha. You don't need to ask what I used to do the processing but I'll tell you anyway: Iris.
I didn't use your bias frames for 2 reasons:
1) They weren't well matched to the lights and so introduced extra banding noise
2) The flats were ISO 100 so using ISO 800 bias frames causes a mismatch - this might also be part of the cause of your banding.
So I used a synthetic flat: all pixels have value 1024 (the correct value for the Canon 450D).
Larger version here:
http://www.markshelley.co.uk/Astronomy/2012/carole_california.jpg
By the way, I noticed quite a few faint fuzzies in the background.
Mark
Thanks Fay, what did you stack it with?
Thank you Mark.
That was interesting what you said about the Bias frames as I always thought that they had to be at the same ISO as the Light and darks, and have always been told that the correct way to do flats was AV 100 ISO. It might certainly explain why there is a problem with the banding and flat calibration if in fact the Bias should also be done at the same ISO as the flats - 100 ISO.
I have never had any success using an artificial flat as I can never get the subtract procedure to work properly in photoshop.
Mark I didn't put any Ha data into Dropbox, but I will if you want to have a "go" at it.
My problem now though is I need to spend time learning to image and process with the Atik, so learning what to do and what not to do with the DSLR is sort of "on the shelf" at the moment.
Thanks for processing the image both of you.
Carole
DSS Carole only DSLR subs, flats bias lights darks
QuoteI have never had any success using an artificial flat as I can never get the subtract procedure to work properly in photoshop
Probably because the subtract only works correctly in 8bit, there are a lot of issues with the 16bit functions, not doing as they should.
Mac.
QuoteProbably because the subtract only works correctly in 8bit, there are a lot of issues with the 16bit functions, not doing as they should.
Hmmm might give that a try, though it would be ideal to subtract the flat at the beginning of processing rather than the end, but to prove a point I might give it a "go".
What version of PS do you use Mac? I have a good reason for asking but don't want to hi-jack my own thread so won't explain here.
Carole
Blimey 11 reprocesses of the same image in one thread....! must be a record.....
btw - Robert's looks the best on my monitor..
QuoteWhat version of PS do you use Mac?
CS4 now
Quote from: JohnP on Oct 30, 2012, 16:26:03
Blimey 11 reprocesses of the same image in one thread....! must be a record.....
Well I'll never need to go there on holiday, as I seem to know it now like the back of my hand. Saved me a fortune! ;)
Tony G