Calibrating colors with background galaxies

[vicent_peris]

Hi,

This is a very short tutorial. We're working now with one of our latest images acquired at CAHA. The photographed object is PK164+31.1, a planetary nebula in Lynx. The original RGB image looks like this:



While thinking about how to calibrate the color of this image, I saw the tiny spiral galaxies at right. So I decided to try to use these galaxies as the white reference of the image. Take into account that these galaxies are significantly redshifted, as well as all the other background objects. So, to me, this would have sense from a documentary point of view. The reference galaxies are these ones:



After applying the color calibration routine, we get this result:



And this is the image after raising the color saturation:



Of course, the H-alpha emission is somewhat muted because we are giving less weight to the red channel. But this can be solved because we have H-alpha data as well. This is a close comparison of the reference galaxies before and after the color calibration process, and after the color saturation curve:



Now these galaxies have blue-cyan spiral arms and orange bulges. See also the we can now appreciate much better the color of the spiral arms of the tiny galaxy inside the PN.


That's all.
Best regards,
Vicent.

[Nocturnal]

I like the result but is removing the red shift not counter the documentary principles? In other words can you reasonably use a non-white light source as your white reference?

[vicent_peris]

I like the result but is removing the red shift not counter the documentary principles? In other words can you reasonably use a non-white light source as your white reference?

Yes, of course. What means white for you?


V.

[Javio]

Spectral agnosticism!!! 

[troypiggo]

I like the resulting color, but am confused b something. Maybe this is what Sander was asking about? With my limited astronomical understanding, the galaxies would be red shifted because they're so far away and moving away?  But the stars and nebulae in the image are not because they're in our galaxy? So calibrating on something that you know is red shifted with much that isn't, makes most of the objects in the image not-quite-right for the sake of some galaxies?

[Nocturnal]

Right Troy, that was my concern too. I would not want to whiten a galaxy I know is red. Instead I'd be happy if the red-shift showed in the final image.

[vicent_peris]

Hi,

It all depends on what you want to do with your image. In this case, I want to calibrate the color with a spiral galaxy which has a redshift of z=0.07. This expands the tone representation through z and, in fact, you can see a lot of small red dots that are in the background. Of course, calibrating with this source we give you somewhat bluer stars.

I've done a similar approach with the ALHAMBRA Survey images and it worked extremely well.


Regards,
Vicent.

[RBA]

It all depends on what you want to do with your image.

Precisely. Color in an astroimage is always relative to something, whether it's a redshifted galaxy, a g2v star, or anything else you can think of. The problem is to think that by using this or that reference you'll get the "correct" (real, natural, etc.) colors.

[ManuelJ]

I'm always amazed by the SNR you can achieve without any effort in professional equipments. This is a 13 hour WIP exposure in Ha. I really don't want to know your exposure, but I'm sure it's super short in comparison

[vicent_peris]

Well... This is about 16 hours of exposure with a Ha filter:




 

Vicent.

[ManuelJ]

Now I'm crying. But that's not a narrow Ha, I'm I wrong?

That signal from everywhere, it cannot be 100% Ha

[RBA]

Did I just see a guy with a real sweet 17" scope and one of the best "somewhat affordable" CCDs in the market  crying??

Coñe... 

[vicent_peris]

Now I'm crying. But that's not a narrow Ha, I'm I wrong?

That signal from everywhere, it cannot be 100% Ha

No, it's a 7 nm Baader H-alpha filter. The red image goes deeper for the background galaxies but not for the outer halo of the nebula.


V.

[ManuelJ]

Now I'm crying. But that's not a narrow Ha, I'm I wrong?

That signal from everywhere, it cannot be 100% Ha

No, it's a 7 nm Baader H-alpha filter. The red image goes deeper for the background galaxies but not for the outer halo of the nebula.


V.

Well, it may say 7nm in the box, but I think it's actually bigger. Do you want a super 3nm filter with a superb transmission? 

RBA: yes, I feel like a child with a toy

[vicent_peris]

Hi again,

I'm acquiring right now fresh data at CAHA. After dinner I will do some photometry to a R and H-alpha images I just acquired because the atmospheric extinction is very stable now. From the brightness ratio between the stars in both images we can derive (approximately) the width of this filter.


Regards,
V.