Lagoon and Trifid - Feedback requested and greatly appreciated

Entropy · #1 02-06-2014, 04:44 PM

Hey everyone,

This is an image of the Lagoon and Trifid nebulas.
23 x 450s subs at 400ISO for around 3hrs total from a Canon 60D with a Takahashi FSQ106 taken at Wiruna. No calibration frames due to a stuff up with the darks and, because of that, laziness with the bias.

This was taken at Wiruna at SPSP14

Full Sized Image

This is my first serious play at Pixinsight, and while I absolutely love this software I am a bit concerned I may have overcooked the image a bit (maybe to much sharpening or playing with colour saturation?)

All criticisms are welcomed... pick it to shreds please

Thanks

rustigsmed (Russell) · #2 02-06-2014, 05:12 PM

Hi Entropy,

Is that link the actual full size image? it looks pretty nice and well framed my comments (i am no means an expert - and have no experience with Pixinsight).

It appears to me that you might have more nebula in there to bring out and you may have overdone the 'highlights' or upped the contrast a tad too far?

Still, very nice though - and i hear Pixinsight has some serious learning curves!

Rusty

rcheshire (Rowland) · #3 02-06-2014, 06:50 PM

Entropy. You have done a good job extracting detail. Given the total time you should see a mass of stars with a good range of colour, particularly at iso400 - good dynamic range. though it looks a little clipped - a bit too dark. With a decent set of darks and bias frames it will be superb.

Amaranthus (Barry) · #4 02-06-2014, 06:57 PM

Did you record the temperature at the time you were imaging? If so, you should be able to recreate your Darks. And the bias can be done any time.

Entropy · #5 02-06-2014, 07:51 PM

Thanks everyone for the critique so far, very helpful as I intend to do this over tomorrow from scratch again.
Barry- recreating the darks right now, so hopefully be able to apply them tomorrow, along with some bias frames... seriously considering buying that lightbox in the classifieds section as well.
Rowland- By clipped, do you mean that when I was playing with the histogram I may have moved the black slider too far, or when I was moving the mid tones slider?
Russell- Not too sure how photobucket works, perhaps I should move to Astrobin, but I think if you click the link then the zoom button it will give full sized image. I was a bit nervous when I was stretching as I thought I was blowing out the core of the nebule, not realising that that gets fixed up in a later step of the workflow, so will be a bit more aggressive with it tomorrow... and a bit less aggressive with the final curves adjustments

.

Thanks again for your comments so far... please keep them coming. I'm very excited by my results today and I am like a sponge at the moment taking in so much information and pointers.

LightningNZ (Cam) · #6 02-06-2014, 10:56 PM

Hi Entropy,

Considering how much exposure you have under dark skies, I think there's likely a lot of data left on the table. The stars are nice and round and the colour in the nebula looks spot on.

Unfortunately is also clearly "black clipped" - the sky is never truly black but has noise in it. It's also been over-smoothed and as a result you've lost colour in your stars and made the background look blotchy and hazy.

I think if you give it a bit of the "less is more" style of treatment you'll have a really wonderful image.

Hope this helps,
Cam

rcheshire (Rowland) · #7 03-06-2014, 05:50 AM

Be careful with curves and clipping pixels with histograms as in some tutorials. I tend not to use the histogram for clipping outlying pixels. DSLR data is highly modified anyway. Personally, I like to leave contrast adjustments until last thing.

Just by way of example, ignoring the colour palette used, this was taken with an unmodded DSLR of similar total time, give or take.
http://www.astrobin.com/full/41162/C/

Entropy · #8 03-06-2014, 04:06 PM

So, I've spent the best part of today playing with Pixinsight and my image to try and improve on it.
While I think I may have improved the black clipping problem, and improved the detail on the outlying areas of the nebula, I was hoping to bring out more nebulosity which is appearing only faintly above and below the Lagoon. I just cant bring it out, any ideas on how I can do this?

BTW I also have added darks and bias frames to this stack.

So what do you think? Am I improving?
Have I ruined the colours (Post stack looked very blue and I think that carried over)?
Am I still clipping data? How can you tell, was it that red blotchy-ness behind the star fields?

Thank you very much for your help so far, it has been great.

Full Size(Now Astrobin'ed)

LightningNZ (Cam) · #9 03-06-2014, 04:54 PM

Okay so you've definitely pulled out more of what was there!

How do we know you're clipping? Well, we don't, but we're guessing. You sky background (between the stars) looks black. It should never be perfectly black. Also check the brightest parts of the nebula or some stars that aren't the very brightest, they should never have RGB values of 255 - which indicates the maximum brightness of an 8-bit pixel colour value.

You can also see a haziness around all the stars - which appear flat, they don't fade off to towards the edges much. This indicates that you've used some form of median filter or smoothing to clean up the noise in the image. Real images have noise. Real views with your eye have noise. Anything that has noise is artistic but it ain't real. Some people love it, other's don't. It's up to you whether you use present your images in this way or not - I'm just giving you my feedback.

Hope that's helpful,
Cam

rustigsmed (Russell) · #10 03-06-2014, 04:56 PM

Quote:

Originally Posted by Entropy

So, I've spent the best part of today playing with Pixinsight and my image to try and improve on it.
While I think I may have improved the black clipping problem, and improved the detail on the outlying areas of the nebula, I was hoping to bring out more nebulosity which is appearing only faintly above and below the Lagoon. I just cant bring it out, any ideas on how I can do this?

BTW I also have added darks and bias frames to this stack.

So what do you think? Am I improving?
Have I ruined the colours (Post stack looked very blue and I think that carried over)?
Am I still clipping data? How can you tell, was it that red blotchy-ness behind the star fields?

Thank you very much for your help so far, it has been great.

Full Size(Now Astrobin'ed)

Hi Entropy,

I definitely think that its a big improvement

Again, I don't know Pixinsight but my gut says it could still go further to bring out the nebulas. But it is hard to know how much without seeing the original tiff file. The blacks still look a bit black to me

Cheers

Rusty

rcheshire (Rowland) · #11 03-06-2014, 05:08 PM

It's coming along nicely. Camera is unmodded from what I can see, so don't expect a lot of red nebulosity. Another source of clipped data can be calibration. This is sometimes noticeable in Pixinsight with DSLR data, if you follow the dark scaling method, which is the PI default.

I suggest the following; use your bias frames to calibrate your flats only - don't calibrate your darks - leave the bias in the dark, if you are not already doing that. Create a master flat and master dark separately and then load up your master dark, master flat and lights in BPP, if you're using that. Give it a go and see if the image will stretch more readily with more colour into the stars. EDIT: You need to create a master bias for the flats.

I don't use pixel rejection when creating master frames because DSLR RAW data is modified in the camera.