Rosette Nebula Image - Help Needed

[bendiddley]

Hey guys. Finally managed a session the other night to photograph the rosette nebula, was delighted to capture it having not managed any photography for a while but I'm a bit underwhelmed by it and looking for some advise on how to improve it. I've used a modified Canon 1100D with an astronomik light pollution filter on a Skywatcher 130PDS. It's made up of 19 subs @ 5mins each so 95mins total, although some thin clouds came in half way through so some of my subs were rather weak. I used 25 darks/bias/flats. I took the flats using a light panel which I placed over the top of the scope. Not sure what else to say about how I took the shots. The image shown is the final stacked image straight out of DSS so no photoshop work done on it. I've played around with it in photoshop but the vignetting and noise just gets really bad. The vignetting is the main thing I'm not happy with. I was expecting the flats to get rid of that, any ideas, have included one of my flats encase that helps. The other thing is I find the colour is rather dull, was expecting something a bit brighter and more red in colour and some colour in the stars too. As said I have tried increasing the red in photoshop but was expecting something a bit more red to start with. How can I achieve that? Is it that the image just needs more exposure time? Hope you guys can help. Cheers.

[AKB]

Well, here is my not very good attempt (I am no expert) but it does show that some improvement (?) is possible.

I ignore the flat – partly because I didn't know which way up to use it, but also because it's rather hard to back out the effects of a poorly applied flat (the flat itself, when stretched, looks fine to me.)

Also, it would perhaps be better to post a non-JPEG version of the stacked image (to ensure that we have full depth) without the flat applied.

All I have done is a little stretching, some luminance masking, and saturation adjustments.

I'm sure others can come along with much better efforts.

 

[almcl]

What stacking/processing software did you use?

I've run the image through StarTools (cropped and binned quite a bit to speed up processing) and there seems to be colour there (see below) although rather less red than I would expect from an astro modded DSLR. 

You have nice, round stars  but as you say, your flats don't seem to be working and you might need to revisit how you are making them.

[bendiddley]

1 minute ago, almcl said:

What stacking/processing software did you use?

I've run the image through StarTools (cropped and binned quite a bit to speed up processing) and there seems to be colour there (see below) although rather less red than I would expect from an astro modded DSLR.

You have nice, round stars but as you say, your flats don't seem to be working and you might need to revisit how you are making them.

Thanks. I used DSS. Re the colour, maybe the cloud cover that affected some of the subs could have played a part in this, in terms of the amount of light that was captured, ie. more light = more colour??

[bendiddley]

47 minutes ago, AKB said:

Well, here is my not very good attempt (I am no expert) but it does show that some improvement (?) is possible.

I ignore the flat – partly because I didn't know which way up to use it, but also because it's rather hard to back out the effects of a poorly applied flat (the flat itself, when stretched, looks fine to me.)

Also, it would perhaps be better to post a non-JPEG version of the stacked image (to ensure that we have full depth) without the flat applied.

All I have done is a little stretching, some luminance masking, and saturation adjustments.

I'm sure others can come along with much better efforts.

 

 

Wow, is that my image? I thought it was one you've done. What is luminance masking? Think I might need to give that a go!

[almcl]

If you are using DSS to post process, you may find switching to something else gives better results.  (Gimp, Photoshop, Astro Pixel Processor, StarTools, Astro Art, Paintshop Pro are a few possibilities.) Also try removing 'cloudy' subs from the list prior to stacking and see if this improves things?  Sorting by Sky background and removing any much-higher-than-average subs might help with the colour saturation.  

[david_taurus83]

Putting the light panel directly over the OTA on a Newtonian is asking for trouble as it can introduce stray light into the focuser and into the light path. Try stacking all again in DSS but leave out the flats and see how it comes out. If the gradients improve then you know where the issue is.

[carastro]

I would have expected much  more from a modified DSLR, SW130PDS and 19 x 5min subs.   the colour is washed out and the flats don;t appear to have worked, so something is wrong here.

Firstly as suggested above, is it possible to identify which subs had clouds through them and remove,  I know it's not easy to see this with unprocessed CR2 files.  

Flats: You say you used a light panel.  Did you do these in the dark or in the daytime?  My 130PDS leals light if I do them in the day time.  If I ever have to resort to doing them in the daytime, I normally cover the primary end of the scope with a black breathable bag.  

The important thing is to get the histogram in roughly the right place, which is between 1/3 and halfway accross from the left.  further over than that it will be over exposed and won;t work.  You will get a much easier to process image once the flats are right.

Is the camera removed from the scope?  If not can you try re-doing the flats?

Carole 

 

[AKB]

55 minutes ago, bendiddley said:

What is luminance masking? Think I might need to give that a go!

I extracted the luminance part of the image, and applied it to the whole image to increase the saturation in the brighter parts of the image, and reduce it in the darker parts to get a nice neutral background without clipping the black levels. 

I happen to use PixInsight as a processing tool, but PaintShop or any number of others can do this. 

[carastro]

With a fair bit of heavy handed processing in Photoshop, I managed to get this.  I compensated for the flats by using gradient exterminator. 

Increased the colour in Match colour several times and selective colour, and HLVG to get rid of the resulting green stars.

I think this shows that the basic data is not too bad once flats are applied correctly, but just wondering what happened to the original colour. 

Could it be there wrong bayer matrix was selected I am not very au fait with this. 

Carole 

 

[bendiddley]

21 minutes ago, david_taurus83 said:

Putting the light panel directly over the OTA on a Newtonian is asking for trouble as it can introduce stray light into the focuser and into the light path.

How should I do it if I don’t put the panel directly over the ota? I currently put it on top of the dew shield, not sure if that makes any difference?

[carastro]

If the light panel is surrounded by a box so the light is only going down the tube it should be OK.   Something like this.  I would remove the dew shield and put the box directly over the aperture.  I have used this method with my SW130PDS and it works OK. 

Did you read what i wrote above regarding the histogram?  If the flat is too bright it won;t work. 

Carole 

[bendiddley]

3 minutes ago, carastro said:

With a fair bit of heavy handed processing in Photoshop, I managed to get this.  I compensated for the flats by using gradient exterminator. 

Increased the colour in Match colour several times and selective colour, and HLVG to get rid of the resulting green stars.

I think this shows that the basic data is not too bad once flats are applied correctly, but just wondering what happened to the colour. 

Carole 

 

Wow thanks Carole that looks amazing. I really need to brush up on my photoshop skills. I may come back to you to ask in more detail on this when I have another go at post processing. I’m pretty good with pshop as im a graphic designer but there’s lots I still need to learn when it comes to processing Astro photos. Yep I’m wondering about the colour too. I’ve always struggled with it in all my Astro photos to date, they’ve always been quite washed out although until this one I had been using an Unmodded camera, this shot is the first time I’ve used a modded dslr. 

[carastro]

Quote

 I really need to brush up on my photoshop skills. I may come back to you to ask in more detail on this when I have another go at post processing. 

I have a few processing tutorials on You Tube, or linked from my website.  Though they are mainly to do with mono imaging, but there will be the odd tip here and there you can use on DSLR images. 

Carole 

[bendiddley]

2 minutes ago, carastro said:

If the light panel is surrounded by a box so the light is only going down the tube it should be OK.   Something like this.  

Did you read what i wrote above regarding the histogram?  If the flat is too bright it won;t work. 

Carole 

Thanks. I use an a3 Huion  light panel. There’s lots of light coming off this outside the ota as it is obviously a lot wider than the ota. Should the light only go over the ota and no wider? Regards the histogram yes I noted your comment. Did you see the flat I posted at the beginning. Does the histogram look ok on that?

[carastro]

I think it would be best to make a simple box like I have to avoid light getting in at the focusser as suggested above.

But there is something very odd with your flats anyway.

Firstly, they are green, and I checked the histogram, and there is virtually no red signal in there at all.  This might explain why your actual image is showing virtually no red signal.  Where was it modified? 

 

The flat itself looks correctly exposed, as stretched it is showing vignetting and dust.  (The nignetting was clearer before I stretched it further to show the dust).

I am no great shakes on the technical stuff, but I am thinking something has gone wrong in the stacking procedure,  I can;t believe that something serious has gone wrong in the modifying procedure. 

Need some input from some-one more techy than me.

Carole 

[bendiddley]

3 minutes ago, carastro said:

I think it would be best to make a simple box like I have to avoid light getting in at the focusser as suggested above.

But there is something very odd with your flats anyway.

Firstly, they are green, and I checked the histogram, and there is virtually no red signal in there at all.

 

That's because I'm using a light pollution filter. Should I not be using one?

[carastro]

If you need a LPF you need one, I can't believe it would knock out almost the entire red channel.  I used to use a Modified DSLR with a CLS filter and whilst the flats were a funny colour, the images were fine.  

Carole 

[carastro]

OK, Ignore what I said about the flats histogram, I think it was a red herring.

The original image histogram was fine.   So I am now thinking more it's to do with the stacking procedure.

Vlaiv where are you, need techy input,.

Carole 

[bendiddley]

1 minute ago, carastro said:

OK, Ignore what I said about the flats histogram, I think it was a red herring.

The original image histogram was fine. So I am now thinking more it's to do with the stacking procedure.

Vlaiv where are you, need techy input,.

Carole

I've always struggled with washed out images and I use the same stacking settings for everything, pretty much, maybe there is something in how I am processing the images which is causing this.

[carastro]

I don't think it is your post processing that is losing the colour, it was already lost before I started to process it.  But I am wondering if it is to do with whatever settings you are choosing in DSS.

I am a bit out of practice stacking coloured data, but if you could upload the files somewhere for us to have a go at and see whether OUR settings produce a different result.  I have the same DSLR as you, so if it works for me, I can tell you what settings I use.

Light,s flats, darks and Bias. 

Carole 

 

 

[bendiddley]

4 minutes ago, carastro said:

I don't think it is your post processing that is losing the colour, it was already lost before I started to process it. But I am wondering if it is to do with whatever settings you are choosing in DSS.

I am a bit out of practice stacking coloured data, but if you could upload the files somewhere for us to have a go at and see whether OUR settings produce a different result. I have the same DSLR as you, so if it works for me, I can tell you what settings I use.

Light,s flats, darks and Bias.

Carole

 

 

Ok great. Not sure where I can upload the files for people to download. I could we transfer them to you personally if you're happy with that?

[carastro]

I've PMd you

Carole 

[carastro]

OK, good news.  There is bags of colour in the image, So I think there must be something wrong with your stacking settings.  Also the flats seem to work OK.   

I'll send you my stacking settings in a PM. 

One thing I never do in DSS is any post-processing at all (I was advised not to), and I don't post-process the Autosave, file I always save my own file.  I have found in the past the results are more inferior in the autosaved file.

So without more ado, Here is the result.  I mainly did levels and curves and colour balance, with a little bit of gradient exterminator as there was a colour gradient in the sky.  I also slightly increased the star colour which is part of my routine these days as the stretching seems to wash out the stars, so I do this at an early stage in the processing BEFORE they get washed out. 

Carole 

 

[bendiddley]

10 hours ago, carastro said:

OK, good news.  There is bags of colour in the image, So I think there must be something wrong with your stacking settings.  Also the flats seem to work OK.   

I'll send you my stacking settings in a PM. 

One thing I never do in DSS is any post-processing at all (I was advised not to), and I don't post-process the Autosave, file I always save my own file.  I have found in the past the results are more inferior in the autosaved file.

So without more ado, Here is the result.  I mainly did levels and curves and colour balance, with a little bit of gradient exterminator as there was a colour gradient in the sky.  I also slightly increased the star colour which is part of my routine these days as the stretching seems to wash out the stars, so I do this at an early stage in the processing BEFORE they get washed out. 

Carole 

 

Holy k'moly! That's amazing!! Can't believe it's the same image. Thank you soooo much. It's incredible to see how much more colour and detail there is and so glad that the colour was there in the files. Can't wait to have another go using your DSS settings. Will send have another crack at it tonight. Hope I can get the same results. thank you!!