Why are my colour images coming out in black and white?

[rojach78]

Hi All,

I'm new to Astrophotography, i'm just trying to find my feet. Please could someone explain why my colour images coming out in black and white with very little colour, i have the following setup and use Photoshop to stretch the image. Thanks.

HEQ5 Pro

UV/IR Filter

ASI533MC Pro

Evostar 80ED

20 x 5Min Exposures at gain 101 on attached example

5min.tif

[vlaiv]

You need to debayer your images as part of your workflow.

What software are you using for stacking and what is your workflow?

 

[rojach78]

6 minutes ago, vlaiv said:

You need to debayer your images as part of your workflow.

What software are you using for stacking and what is your workflow?

 

Thanks for your speedy response, i'm very new to this so please bear this in mind with my answers. I use DSS and i just put the stacked image straight into Photoshop, i then convert it to 16bit, level the colours, stretch using curves and levels until i'm happy with the image.

[Budgie1]

You need to debayer in DSS before heading over the PS.

Open DSS, in the Options menu click on the RAW/FITS DDP Settings and a new window opens.

Click the FITS Files tab and put a tick in the "Monochrome 16bit FITS files are RAW files created by a DSLR or a colour CCD camera". Now you need to make sure the bayer pattern for your camera is correct in the drop-down box. I think RGGB is correct for the ASI533MC, but check to make sure.

You can leave the colour adjustments at 1.0000 and the Bayer Matrix Transformation at Bilinear. Now click the Apply button at the bottom of the window and restack your image and it should be colour.  

Edited April 19, 2021 by Budgie1
Spelling

[rojach78]

13 minutes ago, Budgie1 said:

You need to debayer in DSS before heading over the PS.

Open DSS, in the Options menu click on the RAW/FITS DDP Settings and a new window opens.

Click the FITS Files tab and put a tick in the "Monochrome 16bit FITS files are RAW files created by a DSLR or a colour CCD camera". Now you need to make sure the bayer pattern for your camera is correct in the drop-down box. I think RGGB is correct for the ASI533MC, but check to make sure.

You can leave the colour adjustments at 1.0000 and the Bayer Matrix Transformation at Bilinear. Now click the Apply button at the bottom of the window and restack your image and it should be colour.  

Thankyou, you just saved me from eternal frustration!

[vlaiv]

Excellent information given by @Budgie1

That is only the first step. You'll need to weigh your color information later ( ASI533 is very strong in green  and images tend to have green cast that you must balance out).

If you have issues with pale colors - it could be that DSS messed up things. There is option on background calibration and certain options tend to wash out colors:

If you notice issues - try changing background calibration setting (but do pay attention that sigma clip stacking depends on frames being normalized - and you can get worse results if your LP levels change significantly during session and you don't normalize the background between frames).

[rojach78]

Hi again, i have tried all of the above but i'm still not getting colour in my images? i have attached a dss stack to show the colour channels, shouldn't they be over to the left?

[vlaiv]

13 minutes ago, rojach78 said:

Hi again, i have tried all of the above but i'm still not getting colour in my images? i have attached a dss stack to show the colour channels, shouldn't they be over to the left?

It is usually best not to process data in DSS other than stacking.

You can export your data from DSS in 32bit fits format and then process it in Gimp for example. If you like, you can export data and attach it here and people can give best tips on hot to get the processing right.

[Budgie1]

It does look a bit light and I'd expect to see the histogram colour spikes closer together, but I don't normally look at the stacked image in DSS either, I open it in my processing software to see what I've got.

Have you changed the background calibration, as in @vlaiv post above, and if so what settings are you using?

[reezeh]

On 19/04/2021 at 11:58, vlaiv said:

You need to debayer your images as part of your workflow.

What software are you using for stacking and what is your workflow?

 

I'm glad I'm not an astro imager. I'm turning increasingly Murtaugh with every passing day:

"I'm getting too old for this     "

[rojach78]

Needle.tiff is no calibration, per channel.tiff is as.

Needle.TIF Per channel.TIF rgb calibration.TIF

[Budgie1]

I've had a look at all three and the only one I could do anything with is Needle.TIF and that does seem to have colour in it, along with a nice dust mote.  

The Per Channel and reg calibration images are a lot larger (file size) than the other one. When I stack my images as TIF they are mostly 100mb or less, so I don't know why yours are 4x that size.

Rather then giving incorrect or confusing info, I'm going to let Vlaiv advise on this because his knowledge is vastly superior to mine when it comes to these issues.

Once you get the stacking sorted, the data is good and the stars are nice & tight. Below is what I managed to get from Needle.TIF using PixInsight, although a very basic processing of the image.

[Seelive]

Had a look at the needle.tif (gave up with the other 2, the data seemed well over to the right). Although their were hints of colour in the background noise the main image artefacts just appeared to be greyscale.

[rojach78]

1 hour ago, Budgie1 said:

I've had a look at all three and the only one I could do anything with is Needle.TIF and that does seem to have colour in it, along with a nice dust mote.  

The Per Channel and reg calibration images are a lot larger (file size) than the other one. When I stack my images as TIF they are mostly 100mb or less, so I don't know why yours are 4x that size.

Rather then giving incorrect or confusing info, I'm going to let Vlaiv advise on this because his knowledge is vastly superior to mine when it comes to these issues.

Once you get the stacking sorted, the data is good and the stars are nice & tight. Below is what I managed to get from Needle.TIF using PixInsight, although a very basic processing of the image.

Thanks! That's a lot better than I have achieved in PS, there just doesn't seem to be enough colour in any of my images? Frustrating!!

[vlaiv]

I'm having trouble downloading those two 400mb versions - which should not be nearly that large.

3000x3000x3x4 = ~103mb sounds about right - anything over that is probably x2 drizzled (for some strange reason - one should not drizzle unless under sampled - and even then, it is questionable if it works as expected).

 

[vlaiv]

Something is still not good with stacking.

Two 400mb images have been drizzled for no good reason. You should not turn on drizzling.

Also, when exporting data, don't apply stretch that Deep Sky Stacker makes:

In above image:

- Use 32bit format (that is OK - you did that)

- Prefer Fits file format for linear data (like when posting here on SGL) as fits is standard for astronomical data (it keeps actual values and is easier to diagnose and work with in other software than PhotoShop and Gimp).

- Use compression for TIFF (makes smaller files and nothing is lost - use ZIP) if you want to post TIFF

- Embed but do not apply adjustments

In any case - something is wrong with stacking. What stacking parameters did you use?

Here is histogram of green channel data in Needle.TIF image:

Histogram should not look like that and this implies that something wrong was used for stacking - maybe median stacking instead of sigma clip or similar?

[alacant]

On 19/04/2021 at 12:54, rojach78 said:

very little colour

Hi

Nice shot.

To my eyes at least, the colour looks ok. But I'm the last person to ask about colour;)

Maybe you could try adding flat frames to the mix and/or try cleaning the imaging train?

Cheers

 

Edited April 20, 2021 by alacant

[vlaiv]

29 minutes ago, alacant said:

Maybe you could try adding flat frames to the mix and/or try cleaning the imaging train?

I think that flats were taken and applied and did a good job - except for that one dust particle that moved between lights and flats - which is causing that emboss effect.

 

30 minutes ago, alacant said:

To my eyes at least, the colour looks ok.

I don't know - that is typical star tools coloring of the image. There are no teal stars out there, nor greenish ones.

[rojach78]

13 hours ago, vlaiv said:

Something is still not good with stacking.

Two 400mb images have been drizzled for no good reason. You should not turn on drizzling.

Also, when exporting data, don't apply stretch that Deep Sky Stacker makes:

In above image:

- Use 32bit format (that is OK - you did that)

- Prefer Fits file format for linear data (like when posting here on SGL) as fits is standard for astronomical data (it keeps actual values and is easier to diagnose and work with in other software than PhotoShop and Gimp).

- Use compression for TIFF (makes smaller files and nothing is lost - use ZIP) if you want to post TIFF

- Embed but do not apply adjustments

In any case - something is wrong with stacking. What stacking parameters did you use?

Here is histogram of green channel data in Needle.TIF image:

Histogram should not look like that and this implies that something wrong was used for stacking - maybe median stacking instead of sigma clip or similar?

Thanks for this info, using your "Save as" settings has definately re-instated my colours, there does seem to be more green in my images but isn't that normal in these cameras?

NGC4565- 28 x 5 min,101g.TIF

[rojach78]

I have added flats and dark flats but there is a huge dust spot that you are refering too, i will be cleaning before i image again

[rojach78]

15 hours ago, Budgie1 said:

I've had a look at all three and the only one I could do anything with is Needle.TIF and that does seem to have colour in it, along with a nice dust mote.  

The Per Channel and reg calibration images are a lot larger (file size) than the other one. When I stack my images as TIF they are mostly 100mb or less, so I don't know why yours are 4x that size.

Rather then giving incorrect or confusing info, I'm going to let Vlaiv advise on this because his knowledge is vastly superior to mine when it comes to these issues.

Once you get the stacking sorted, the data is good and the stars are nice & tight. Below is what I managed to get from Needle.TIF using PixInsight, although a very basic processing of the image.

Thanks for this, i now think i have fixed the issue, i have uploaded a new stack if you want to take a look, if not then thanks for your help, much appreciated!

[alacant]

10 hours ago, vlaiv said:

typical star tools coloring

StarTools? I didn't use StarTools. Sorry if it's not perfectly balanced. I raise my hand. I'm hopeless at that!

Cheers

[rojach78]

11 hours ago, alacant said:

Hi

Nice shot.

To my eyes at least, the colour looks ok. But I'm the last person to ask about colour;)

Maybe you could try adding flat frames to the mix and/or try cleaning the imaging train?

Cheers

 

Thanks for this, i think i have corrected the issue thanks to 

 

10 hours ago, vlaiv said:

I think that flats were taken and applied and did a good job - except for that one dust particle that moved between lights and flats - which is causing that emboss effect.

 

I don't know - that is typical star tools coloring of the image. There are no teal stars out there, nor greenish ones.

 

11 hours ago, alacant said:

Hi

Nice shot.

To my eyes at least, the colour looks ok. But I'm the last person to ask about colour;)

Maybe you could try adding flat frames to the mix and/or try cleaning the imaging train?

Cheers

 

Thanks for this, i think i have corrected the issue thanks to Vlaiv's "save as" settings. I have applied Flats and Darks but there i a huge dust spot which i am going to clean before i next image!

[vlaiv]

52 minutes ago, alacant said:

StarTools? I didn't use StarTools. Sorry if it's not perfectly balanced. I raise my hand. I'm hopeless at that!

Cheers

That is interesting - auto color balance in StarTools usually produces that sort of colors, so I assumed you used that.

In fact - I've seen what I believed to be StarTools processing in many of your images. I was obviously wrong, but it is interesting none the less.

[vlaiv]

1 hour ago, rojach78 said:

Thanks for this info, using your "Save as" settings has definately re-instated my colours, there does seem to be more green in my images but isn't that normal in these cameras?

Yes that is quite normal for raw data and it needs to be corrected.

Simplest way is to assign weights to each channel. You can find F3 class star and make sure it is white - measure pixel values and scale each channel so that F3 class star is white.