Opinions on this Colour palet (modified SHO).

[Adam J]

This image is not complete yet, still got more data to capture.

I guess I have never liked the more traditional Gold / Blue SHO style, not least because I dont understand why you would want to collect three channels and process it into an image containing only two main colours.

Hence, this is what I have come up with. I like it but wondered what other people think of it. One of the things that has driven me in this direction is the lack of OIII in comparision to SII signal in this object hence it starts to look almost mono chromatic gold and very little blue in the traditional processing.

 

Let me know yout thoughts.

Adam

 

Edited September 12, 2020 by Adam J

[retret66]

We have similar style and I love the yellow to cyan transition.  We both can agree that when we image 3 different channels, we should get 3 different colors to separate..

[MarkAR]

Love the fiery colour scheme.

[andrew s]

I have no problems with and quite like your colours, however, I don't  understand why, for narrow band in the visual, people don't just use the colour associated with the wavelength.

Regards Andrew 

[vlaiv]

16 minutes ago, andrew s said:

I have no problems with and quite like your colours, however, I don't  understand why, for narrow band in the visual, people don't just use the colour associated with the wavelength.

Regards Andrew 

Often they do - except is not one of three primary colors R, G or B, but some other primary colors. No particular reason to choose R, G and B primaries of sRGB color system for rendering narrowband images except for simplicity,

If one wants to be very specific about it - they can devise color transform between their primaries and RGB color space for consistent results.

In fact, choice of particular primaries can lead to aesthetically very pleasing results. This gives me brilliant idea - software that will specify primaries by using color picker and doing all the necessary color processing on NB image - cool

[andrew s]

@vlaiv  if I understand you correctly, I am not sure they would necessarily be primary colours. For example H alpha would be a specific shade of red and there could be more than 3 NB filters. They aim would be to present the image as we would see it if it were bright enough to trigger out cones.

Regards Andrew 

[R26 oldtimer]

That looks very pleasing!

[vlaiv]

24 minutes ago, andrew s said:

@vlaiv  if I understand you correctly, I am not sure they would necessarily be primary colours. For example H alpha would be a specific shade of red and there could be more than 3 NB filters. They aim would be to present the image as we would see it if it were bright enough to trigger out cones.

Regards Andrew 

Oh, ok. I misunderstood your post above. I was under impression you were talking about "color associated with wavelength" in sense of SHO = RGB

But what you are suggesting is to do natural color mapping.

There are couple of issues with that - why it is not useful.

1. No screen can display individual wavelengths of light. It can't be done with trichromatic color system. You can't display color associated with 656nm wavelength with regular displays. You would need display with 3 types of lights each corresponding with particular light to be able to display properly such image - all other screens would do approximation.

This is XY chromaticity diagram with different trichromatic systems. Vertices of triangles represent particular primary color and such system can only show colors inside the triangle. Red triangle that encompasses all colors has "virtual" primaries - such colors don't exist outside of mathematical models and can't be reproduced. Black line that runs on the edge of color region is composed out of individual wavelengths.

Givne the above it is clear that only way to have three pure wavelengths covered with a triangle is to put vertices in exact points on outer edge or to have virtual primaries (meaning no display possible).

2. Even if above was possible SII and Ha are so close in wavelength that we could have very hard time telling them apart even at same intensity let alone in varying intensity of image.

Above is rather interesting as it shows that natural rendition of SII, Ha and OIII image for example would always have these shades:

Only shades below black line that joins 656/672 and 500nm. Very narrow gamut of hues.

[wimvb]

Regardless of colour theory, I also like the fiery look of this palette