Jump to content

Narrowband

Colors in M42 core


Recommended Posts

Hello there,

I recently took some images of the Orion Nebula using my DSLR and SW 150 PDS. Not sure if I went for the right settings, but I use 8 seconds exposure time and ISO 800. Then I stacked all my images together (14 min total) and played with it in GIMP.

I am actually quite happy with the end result (as a beginner, mind you), but there is something puzzling me: when looking at one single frame, the core of the nebula has a blueish color to it, while in the stacked and stretched  image the core has more of a pink hue (see image below comparing a single frame, on the left, with my final image).  I actually like the blue core, as it makes the image a bit eerie. But now I wonder why it doesn’t look like that in the final image… Could this be the result of me stretching the image and somehow messing up with the colors?

Thanks in advance!

 

Orion core.PNG

  • Like 1
Link to comment
Share on other sites

For a first image that it fantastic.
The loss of colour is just a factor of doing the stretching. There are ways in some software to enhance certain colours selectively in some software to preserve them, to a degree, but generally as you brighten things up the brighter areas do loose their deep dark colours.
Also ways of masking certain bright areas as you stretch so it stretches the dark areas but not the masked parts, not sure how to do in Gimp though or if that is possible. But processing is a whole new world and getting the data is only half the story.

Keep it up though there should be some great images to come if that is anything to go by 🙂 

Steve

Edited by teoria_del_big_bang
  • Thanks 1
Link to comment
Share on other sites

Yes, stretching tends to desaturate colors.

It is non linear transformation that can be be approximated with power law for example.

Let's take very nice color that has ratio of 4:2:1 of R:G:B components, but make it very dark by scaling it to be say

0.04, 0.02 and 0.01 in 0-1 range

Important thing to note is that red is 4 times as strong as blue and green is two times as strong as blue (and consequently - red is twice as strong as green).

Now we can "stretch" our data by applying power law - say raise values to power of 0.1.

This is the same as curves that look like this:

image.png.d294d36a0f2411cbd20557f4881f89db.png

In any case, resulting triplet will be:

0.04^0.1, 0.02^0.1 and 0.01^0.1 or when we calculate values:

~0.72478, ~0.67624 and ~0.63096

Not only that values are no longer 4:2:1 but they are quite close in value all being in range 0.63 - 0.73. If you take such color - it will be very grey and washed out - desaturated as all three components are now very close in value.

In fact - if we look at actual colors (say 0.8, 0.4 and 0.2 and above tripled of 0.72, 0.68 and 0.63) - here is what we get:

image.png.ddcc7830331e40fc64b12f6e689085e7.png

left is 4:2:1 RGB ratio and right is after stretching or 0.72 : 0.68 : 0.63 ratio (we can see that it is still sort of the same color - but greatly desaturated).

If you want to preserve color you have to stretch luminance only and apply calculated RGB ratios for each pixel to final stretched luminance.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, Denys said:

when looking at one single frame, the core of the nebula has a blueish color to it

What is the colour of the core in the stacked but not stretched image? Does the colour change when you do colour calibration or when you remove any light pollution?

What Steve ( @teoria_del_big_bang) is referring to is called masked stretch. In essence, you apply a mild stretch to the image, extract luminance and use that inverted as a mask to the image when you apply a second mild stretch. This can be repeated a couple of times.

Pixinsight has a process called masked stretch, which applies this procedure up to 100 times to the image. It keeps colour in the bright areas but at the cost of producing a flat image.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Thanks for your replies, I really appreciate!

56 minutes ago, teoria_del_big_bang said:

Also ways of masking certain bright areas as you stretch so it stretches the dark areas but not the masked parts

I did not know about this, but will surely look into it!

56 minutes ago, vlaiv said:

if we look at actual colors (say 0.8, 0.4 and 0.2 and above tripled of 0.72, 0.68 and 0.63) - here is what we get:

Thanks for the very illustrative example vlaiv!

23 minutes ago, wimvb said:

What is the colour of the core in the stacked but not stretched image? Does the colour change when you do colour calibration or when you remove any light pollution?

Oh, should have checked the color of the stacked only image before, it is actually closer to pink... I did not try to remove light pollution, as I was told I am under Bortle 4 sky (according to light pollution map anyway), so hopefully that means I can get away without this step?

Denys

  • Like 1
Link to comment
Share on other sites

42 minutes ago, Denys said:

so hopefully that means I can get away without this step?

Even in Bortle 4 (and Bortle 2 for that matter) you still have "light pollution", just not man made. Recently we've had spectacular light pollution, aka Aurora Borealis. 😁

In all seriousness, very weak Aurora acts as light pollution and can give a green cast to images.

You also have the sensitivity curve of your camera to consider. Most sensors are more sensitive to green than to blue, and will never give a truely neutral background.

Btw, here's an old article showing how to apply masked stretch in PhotoShop/Gimp (p 86)

https://rasc.ca/sites/default/files/publications/JRASC-2012-04-hr.pdf

https://rasc.ca/jrasc-2012-04

  • Like 2
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.