• Welcome to Orpington Astronomical Society.
 

News:

New version SMF 2.1.4 installed. You may need to clear cookies and login again...

Main Menu

Iris Nebula and neighbouring dust - Canon EOS R

Started by MarkS, Sep 13, 2021, 23:14:10

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

MarkS

Another image from my (unmodified) Canon EOS R.  It's a target that has defeated me for ages - the Iris Nebula and the nearby clouds of dust.

Acquisition details are the following:
* Unmodified Canon EOS R on Takahashi Epsilon 180ED
* 60 x 2min dithered exposures at ISO 1600
* Sky quality reading averaged 20.80



Larger version here:
http://www.markshelley.co.uk/Astronomy/2021/iris_canon_eosr_20210906.jpg

Extracting those faint dust clouds is a big challenge with only 2 hours of data, even with my fairly dark sky.  I have attempted the same area of sky with the Sony A7S, the Nikon Z6 and now the Canon EOS R.  Without a shadow of a doubt, the processing of the Canon EOS R data was far easier than the other two cameras.  The Sony A7S and Nikon Z6 both introduced difficult-to-manage colour gradients caused by the internal "cooking" of the raw data.

This image has made the decision process really simple - the camera is a keeper!

Mark 





Carole

#1
Really lovely Mark, and co-incidentally I did exactly the same image a week ago in the Cotswolds, except stupid me forgot to put the cooling on, so it's horribly noisy.  Going to do a home process on it today or tomorrow so will post it up warts and all. 

Mine was done with my CCD camera, so took hours of data.

I always admire your star control and colour and I agree getting the colour right on those dust clouds was tricky.  I intend to get some further COOLED data on this at kelling if the skies are clear.

Carole

MarkS

#2
In addition I used an advanced technique for taking flats.  The colour and brightness of my flats recorded by the sensor matches the colour and brightness of the background sky in the long exposure lights.  It's a technique I have been slowly developing over the past few years.  The flat needs to be taken using a computer screen, so the colour of the display can be altered to exactly match the local light pollution. The screen brightness and exposure length is then altered so the histogram of the flats is in the same place as the histogram of the lights.  It's an iterative process to get this exactly right.

This new style of taking flats simultaneously solves 2 problems:
* Non-linearity in the response curve of the sensor
* Colour cross-talk at the pixel level on one-shot-colour sensors

I often see advice on forums that flats should be taken so the histogram peak is at the halfway point, i.e. so the data are on the linear part of the response curve.  Unfortunately this advice, although well meaning, is not the whole truth.  Certainly the response tends to become non-linear as the sensor nears saturation but I have plenty of evidence to show that the response curve at the halfway point is still far from being sufficiently linear.

Mark

Roberto

Beautiful image and interesting Flat technique.  👍😎  I can see the benefits for OSC cameras.  Can you already see a difference in the calibrated light not being colour biased (typically green)?  What other benefits are there?

Roberto

Carole

Your flats technique sounds great but far too advanced for me to get my head around.

I am just about to post up my version of the same targets. 

Carole

garrick

Great image. Had a look at the larger version and the detail is great. How did you pull the detail out of the dust clouds with 2 hours? For 2 hours of data this is a fantastic image, star colours are impressive.

Clear Skies Mark.

PS I am really lazy with flats, but your technique has intrigued me. Sounds extremely interesting.


MarkS

Quote from: garrick
Great image. Had a look at the larger version and the detail is great. How did you pull the detail out of the dust clouds with 2 hours? For 2 hours of data this is a fantastic image, star colours are impressive.

Thanks.

I think the keys are the following:
* I have a reasonably dark sky - typically SQM 20.8-20.9
* I'm using fast optics ( f/2.8 ) with a large sensor so my light collection is very good
* I've binned the result down to a smaller size to improve signal-to-noise

Mark

RobertM

Another great image Mark, must feel like you've got over a very large hurdle with that Canon.

Robert