Brown m31

[TiffsAndAstro]

My best m31 image seems very brown. I'm hoping this would improve as I add more time to it? Or is it because it's low in the sky? Or me just being bad?

Others' images seem silvery blue but if I turn down saturation mine just goes monochrome ish.

Could a low off set effect this?

Edited July 5 by TiffsAndAstro

[michael.h.f.wilkinson]

Sometimes this is a matter of colour balance. In Astro Pixel Processor there is a tool to calibrate the star colours so their colour distribution follows some expected distribution. That often sorts this out.

[TiffsAndAstro]

17 minutes ago, michael.h.f.wilkinson said:

Sometimes this is a matter of colour balance. In Astro Pixel Processor there is a tool to calibrate the star colours so their colour distribution follows some expected distribution. That often sorts this out.

i believe you. i've just really really unsaturated it across the board and it might look better. thanks for this, i don't have app but i will try something (maybe) similar in siril 

[Elp]

Are your histogram peaks balanced?

When I last did this target most of my stars were yellow/brown but the galaxy looked fine, I suspect some people process it to a cooler colour.

In Siril you can do a photometric colour calibration on linear data, enter the target m31, find, if found it should load at least one entry into the list, check your image pixel scales (it usually populates this automatically based on the fits data), then apply. I've found it sometimes makes the image too cool, but if you look at the histogram before and after (see it in histogram stretch dialogue) you'll see the rgb channel peaks are now aligned.

[michael.h.f.wilkinson]

28 minutes ago, TiffsAndAstro said:

i believe you. i've just really really unsaturated it across the board and it might look better. thanks for this, i don't have app but i will try something (maybe) similar in siril 

People tend to increase saturation in many cases (I do it too), but you have to be careful not to explode colour noise. Here is an example of the same shot, without star colour calibration

and with star colour calibration and some gentle enhancement of saturation

This is about 3.5 hours of data from a Bortle 4-5 site, modded Canon EOS 550D, using an APM 80mm F/6 triplet with Tele-Vue TRF-2008 0.8x reducer. Yes, it needs (much) more data.

[TiffsAndAstro]

5 minutes ago, michael.h.f.wilkinson said:

People tend to increase saturation in many cases (I do it too), but you have to be careful not to explode colour noise. Here is an example of the same shot, without star colour calibration

and with star colour calibration and some gentle enhancement of saturation

This is about 3.5 hours of data from a Bortle 4-5 site, modded Canon EOS 550D, using an APM 80mm F/6 triplet with Tele-Vue TRF-2008 0.8x reducer. Yes, it needs (much) more data.

all you did was use "star colour calibration and some gentle enhancement of saturation"? ffs i don't want to have to learn APP too  

i wasn't even referring to my brown stars  ill try a photometric colour calibration in siril see if that has a similar effect to what you got. my image is surprisingly similar to your first one, so my result is not terribly unusual or anything? thanks for sharing this btw. and yes i need more time, preferably later in the year...

[Clarkey]

9 minutes ago, TiffsAndAstro said:

all you did was use "star colour calibration and some gentle enhancement of saturation"? ffs i don't want to have to learn APP too  

i wasn't even referring to my brown stars  ill try a photometric colour calibration in siril see if that has a similar effect to what you got. my image is surprisingly similar to your first one, so my result is not terribly unusual or anything? thanks for sharing this btw. and yes i need more time, preferably later in the year...

As has been stated many times, the data collection is the 'easy' bit. (Weather / seeing / deep pockets permitting). Getting the best from the data is where the real skill is. I have been doing serious AP for 4 years and I only consider myself a 'competent' amateur.

[TiffsAndAstro]

36 minutes ago, Clarkey said:

As has been stated many times, the data collection is the 'easy' bit. (Weather / seeing / deep pockets permitting). Getting the best from the data is where the real skill is. I have been doing serious AP for 4 years and I only consider myself a 'competent' amateur.

well you made me decide to try harder, and i decided to randomly try photometric colour calibration AND (non photometric) colour calibration and background neutralisation in siril and seem to be getting somewhere. of course, it could all result in a hot mess of pixels but it at least looks more the colour i would expect. maybe

btw £460 for the sc605cc and a free nb filter sounds a really good deal. and i thought my deal was pretty good

Edited July 5 by TiffsAndAstro

[Clarkey]

25 minutes ago, TiffsAndAstro said:

btw £460 for the sc605cc and a free nb filter sounds a really good deal. and i thought my deal was pretty good

Yes, but I might have to pay extra tax if I'm unlucky. Also - I have to rely on support from China.

[TiffsAndAstro]

30 minutes ago, Clarkey said:

Yes, but I might have to pay extra tax if I'm unlucky. Also - I have to rely on support from China.

I paid the £2 extra at checkout for all tax included. Royal mail delivered it, no additional charges.

Support from china is not ideal, but svbony got back to me with questions via email quick enough.

If you're Nina based, I'd be interested in your offset. I used Cuiv's method and settled on 5, not 50 which seems common for other 533 based cameras from other brands.

[TiffsAndAstro]

this shows me its heading the right direction. more time on target it is, i think. bit more drama, less "smoke from forest fire" discolouration

 

 

 

[TiffsAndAstro]

ok final flogging of the horse with this one. my latest result is bluer but has some cartoon binocular gradients like a lobe around the right and left half of m31. unless they're stars? anyway, hopefully that will be reduced with more time on target. about 38x60sec one panel and 44x60sec other panel. 

 

[WolfieGlos]

You've got better colour out of it now, nice job. Just need to work on the background  

Some food for thought on the OP, just on the front of colours and brown, are you stacking in Siril? I found exactly the same when it came to galaxies and got very tired of it.

I came across this topic on Siril's forums, which seemed to collaborate my findings with Siril with galaxies and DSLR's. I don't know if the same happens with astrocams, but in anycase, I followed the suggestion in there to use ASTAP for stacking (and then process in Siril) and I was able to get the blues to really come through. The results posted there are exactly the same as what I experienced, the same hues and colours. Here are some links here on SGL where I posted M63 and M33.

Do not ask me why ASTAP brings out more colour, I don't have that level of understanding; only that I tried it and it provided a much better result. After stacking in ASTAP, I found that when you then open it in Siril, after the linear work, a large AsinH boost, and possibly a modified Asinh stretch through the GHS tool, and the colours would come through.

I still stack with Siril but I now fully process in PI, and now I do not suffer with this problem thanks to the SPCC tool in PI. You can see the difference in M33 in my final post in that thread. I can only assume it's to do with colour calibration in Siril, and I would suffer with the problem both with or without the use of PCC. Oddly it never posed the same problem for star clusters or nebulae.

Colour is very subjective anyway, so let's not open that can of worms.

HTH.

[Elp]

The thing is if you've used the same data and different software gives a different result, the data is still in there. You have to balance r g b to suit the look you want. This is why I always state learning manual processing is better than following astro stuff blindly, you're not learning properly how to edit an image. With good manual image editing skills you can edit any image, not just astro and apply it to other things. AP editing is overall subjective anyway, unless you're in front of the object with direct visual views of it with it illuminated in full colour spectrum light (also illuminating light temperature and intensity can slew perceived colour coming off an object, you can try this yourself indoors with different lighting) it could appear any colour, people also perceive colours slightly differently, sometimes significantly.

I find PCC flattens the colour, and a lot of curves editing, saturation, contrast and selective colour editing is required to make AP images pop, otherwise they look flat and lifeless.

If you're editing on a non colour calibrated monitor again this can affect results, seeing the image on another device again different result, different panel technology again a different result. I can do the same operations on the same data and get a different result. So I tend to take time, it takes hours to fine tune an image, go away and come back again, view it on different devices at different screen brightnesses, adjust and repeat.

Its best to save final editing until you've spent some time acquiring data, by all means stack and do a quick check to see how it's coming along. Through experience you'll know when it is a good time to start editing the final image depending on how many hours you've captured, unless you're imaging difficult objects.

Edited July 6 by Elp

[TiffsAndAstro]

This is very little data, I was only bothering with it so much because it's been cloudy.

I was worried about the colour, but not so much now. Just need to test I can do this with two sessions.

Ty for the help I think colour correction and photometric colour correction in siril achieved at least a measure of what you recommendend.

As for the background, hopefully more time on it and it might improve. Maybe. 

Edited July 6 by TiffsAndAstro

[Elp]

Usually increased time increases the detail to a point (especially for very faint signal), but contributes to noise reduction in both signal and background, more time, smoother result but you'll still need to apply NR to any image you do, I learned this early on.

[TiffsAndAstro]

10 minutes ago, Elp said:

Usually increased time increases the detail to a point (especially for very faint signal), but contributes to noise reduction in both signal and background, more time, smoother result but you'll still need to apply NR to any image you do, I learned this early on.

I'm aiming for 10 hours total. Might need more though. 

[Elp]

Because it's a bright target you can probably get a result in half that time, my last attempt was just over 2 hours but imaging at F3.

[Earl]

It's hard to get a good colour in the outer arms without blowing the core, he brownish tinge from what I recall is pretty much how it is.

[TiffsAndAstro]

2 hours ago, Elp said:

 

Thanks both. Will see what I get with five hours. 

Edited July 6 by TiffsAndAstro

[ollypenrice]

 

One of the biggest aids to setting colour balance is the background sky - if there is any - and in this case there is. There's a tiny bit of IFN and some wispy Ha for those with extremely deep data but , for most people, the surrounding sky is neutral grey. (Equal values in all three colour channels.) This makes a good reference point.

As a general rule imagers exaggerate the blue component in spiral arms. Have a look at the Hubble team's spiral galaxies to see how subdued spiral colour is in their work.

Olly

[900SL]

Has anybody said 'it's brown because it is [removed word]' yet?  

Only joking Tiff! You're rapidly becoming a legend round here and I enjoy reading your latest missives on your journey to the stars

 

[TiffsAndAstro]

2 hours ago, 900SL said:

Has anybody said 'it's brown because it is [removed word]' yet?  

Only joking Tiff! You're rapidly becoming a legend round here and I enjoy reading your latest missives on your journey to the stars

 

Ty for your kind words

I'm trying my best not to look at others' images of my targets until I at least have a go myself. Nina's framing assistant ones don't count.

I thought it looked ok in brown for a little while.....

I managed another 90 ish minutes before rain last night, stacked them ok (there is a slight seam in histogram view) but graxpert background extraction seems to do a good job on the stitched result of both nights, I think it's removing some detail. Need to experiment a bit more before I can put another less brown version on here.

[TiffsAndAstro]

17 hours ago, ollypenrice said:

 

One of the biggest aids to setting colour balance is the background sky - if there is any - and in this case there is. There's a tiny bit of IFN and some wispy Ha for those with extremely deep data but , for most people, the surrounding sky is neutral grey. (Equal values in all three colour channels.) This makes a good reference point.

As a general rule imagers exaggerate the blue component in spiral arms. Have a look at the Hubble team's spiral galaxies to see how subdued spiral colour is in their work.

Olly

I'm finding background neutralisation and non photometric colour calibration helps. I still use photometric too.

Graxpert seems a hindrance at the moment it seems to some remove detail during background extraction and also does some kind of weird stretch and green tint when saving out. Also a bit crashy.

Edited July 7 by TiffsAndAstro

[TiffsAndAstro]

couldnt resist seeing if i could do a 4 panel mosaic of veil nebula. seems i can, but probably shouldn't. will try stacking again but with matching number of exposures. and maybe get more than 30mins per panel. had to bin 2x2 because pc blue screens when running this through star net at 6000 x 6000 pixels

Edited July 8 by TiffsAndAstro