It is currently Tue, 30-05-17, 9:23 GMT

All times are UTC




Post new topic Reply to topic  [ 16 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Mon, 10-09-12, 14:33 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
Let me briefly report about a relatively recent feature of celestia.Sci's spiral and (dwarf) irregular galaxies:

In agreement with observations, I have added a distribution of starforming regions along the spiral arms and randomly in irregular galaxies. These regions are visible as small pink patches of ionized atomic hydrogen (H II) in practically all spirals and irregulars. Hubble images with their great spacial resolution provide striking examples, like M51 (Whirlpool) with its compagnon, displaying a particularly rich distribution of these characteristic pink patches!

Image

H II regions can be seen out to considerable distances in the universe, and the study of extragalactic H II regions is important in determining the distance and chemical composition of other galaxies. Spiral and irregular galaxies contain a lot of H II regions, while elliptical galaxies are almost devoid of them.

In HII regions, the dominant spectral line has a wavelength of 656.3 nm. This is the well-known pink H-alpha line emitted by atomic hydrogen.

Here is now my recent addition of HII regions e.g. for the great M 101 spiral in celestia.Sci (using the SDSS color profile):

[Click on image by all means!]
Image

Here is the view from a distance, displaying this lonely grand spiral in space:
Image

In (dwarf) irregular galaxies, the H II regions can be even much more copious and extended. See e.g. the irregular dwarf galaxy NGC 4449 (Caldwell-21) in the LMC

Image

Of course the pink H II patches are graphically exaggerated to make them most obvious. In celestia.Sci, they are less enhanced.

Fridger


Last edited by t00fri on Mon, 17-09-12, 17:28 GMT, edited 5 times in total.

Top
 Profile  
 
 Post subject:
PostPosted: Mon, 10-09-12, 16:49 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2726
Location: France, South, not far from Montpellier
Pretty nice; would you post the Ir to have a look?

Do you already have in mind the modus operandi to switch between different spectrum? Hard switches or smooth transitions?

(A detail, to my neophyte eye, your sc template is too sharp, need some fluffy stuff all over it! Anyway this is just candy eye, do not lose your time with such tuning, if one day I put a hand on an osX build, I'll tune this for you... :wink: )


Top
 Profile  
 
 Post subject:
PostPosted: Mon, 10-09-12, 19:13 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
ElChristou wrote:
Pretty nice; would you post the Ir to have a look?

I have always been bad with recognizing acronyms ;-) What is "lr"??
Quote:
Do you already have in mind the modus operandi to switch between different spectrum? Hard switches or smooth transitions?

The code and the final strategy are still work in progress. Some aspects I have decided about others not yet.
Quote:
(A detail, to my neophyte eye, your sc template is too sharp, need some fluffy stuff all over it! Anyway this is just candy eye, do not lose your time with such tuning, if one day I put a hand on an osX build, I'll tune this for you... :wink: )


Right. I have many Sc templates to choose from ;-) . But see my more detailed comments below.

Fridger


Last edited by t00fri on Tue, 11-09-12, 8:51 GMT, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Tue, 11-09-12, 8:50 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
Christophe,

one explanation why I had finally given preference in celestia.Sci to spiral templates with practically no noise in-between the arms: While the noise looks more natural in full view onto the disc, statistically most views refer to inclined discs.

In the latter predominant case, for increasingly sideways perspectives the noise amplifies and most spiral structures tend to get wiped out! Hence the price for a little noise in "head-on" view seemed too high.
Despite many graphics experiments I have never come up with a satisfactory solution yet.

In principle, the issue could be improved with a higher template resolution e.g. 256x256, 512x512 instead of only 128x128. But unfortunately and understandably, the frame rates drop dramatically for bigger templates.



Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Tue, 11-09-12, 13:27 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2726
Location: France, South, not far from Montpellier
t00fri wrote:
ElChristou wrote:
Pretty nice; would you post the Ir to have a look?

I have always been bad with recognizing acronyms ;-) What is "lr"??


My bad an r is missing!!
Read Irr for irregular, wanted to have a look how they looks with HII regions enhanced...

Concerning the templates, computers being always more powerful perhaps you should consider a 256 template as default now? (If I recall right, on my old Celestia config I was using a 256 for the MilkyWay... that was 2/3 years ago? Don't remember! :oops: )


Top
 Profile  
 
 Post subject:
PostPosted: Tue, 11-09-12, 21:22 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
ElChristou wrote:
t00fri wrote:
ElChristou wrote:
Pretty nice; would you post the Ir to have a look?

I have always been bad with recognizing acronyms ;-) What is "lr"??


My bad an r is missing!!
Read Irr for irregular, wanted to have a look how they looks with HII regions enhanced...

Concerning the templates, computers being always more powerful perhaps you should consider a 256 template as default now? (If I recall right, on my old Celestia config I was using a 256 for the MilkyWay... that was 2/3 years ago? Don't remember! :oops: )


Wow! No chance for guessing this one ;-)

Of course, I'll put some (dwarf) irregulars with many H II regions together tomorrow. Today, there was too little time left over.

I'll also further investigate the option of 256x256 default sizes for templates.

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Wed, 12-09-12, 16:36 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
Let me discuss next some rich HII starforming occurrences in (dwarf) irregular galaxies
Before doing so, some more general remarks are adequate:

As mentioned above, the H-alpha spectral line that indicates the presence of starforming regions has a wavelength of 656.3 nm. In the standard basis with visual blue, green and red filters, this wavelength corresponds to a conspicuous pink color.

However, some 3 color imaging, notably from SDSS, uses instead a green, red and near infrared filter basis mapped onto the visible blue, green, red colors.

Effectively this turns the pink 656,3 nm line into a conspicuous bright cyan blue color!! (confusing??) . This can be discussed more mathematically, of course.

Note, color photos can vary quite a bit, even if coming from the same (Hubble) address. Besides the NGC4449 image above,

http://forum.celestialmatters.org/userp ... ions_1.jpg,

here are two Hubble alternative images of the dwarf irregular galaxy NGC 4449 with conspicuous pink H II patches:

[Click on images by all means!]
Image
Image

Apparently, among these NGC 4449 images, the displayed differences in colors and the intensity of the H II regions are quite dramatic. Therefore finding an objective criterion of relating different color shadings is essential for a sensible visualization of colors. Color profiles are supposed to be general, flexible and quantitative tools, but due to the complexity of the matter still represent work under development.

Since we cannot compete with a Hubble type resolution, one approach for comparing with celestia.Sci consists in blurring the photographic galaxy images to a matching degree. Along these lines, the first NGC4449 image above, http://forum.celestialmatters.org/userp ... ions_1.jpg
was rotated and blurred with a Gaussian blur of 35 pixel and then reduced in size => image on the right below):

Image

In this qualitative comparison with NGC4449 from celestia.Sci (left) you can see a rough correspondence of the red HII patches in a statistical sense. Much depends here, however, on the amount of graphical HII enhancement that the Hubble people have implemented. See also the alternative images at the beginning of this post.

But there is more: While in the celestia.Sci image (left) I used the SDSS color profile, I rather chose the visual pink color for the HII regions! For SDSS, based on green, red and near infrared filters, the consistent choice would have been the bright blue HII color! Of course the HII colors may be flipped with a click. The natural pink color can be better seen, of course... Here I am still experimenting.

Finally, a couple of SDSS images with correct bright blue for HII starforming regions:

1) M51 with its starforming HII along the arms now in blue instead of pink (see Hubble image above! http://forum.celestialmatters.org/userp ... arge_1.jpg)
Image

2) NGC 2403
Image

++++++++++++++++++++++++++++++++++
I hope that in this discussion the eminent importance of a general color profile framework for sensible comparisons of color in the Universe became apparent
++++++++++++++++++++++++++++++++++

More soon,

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Thu, 13-09-12, 19:29 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
Let me continue a bit with H II regions notably as identifiable in the vast amount of SDSS imaging data.

Actually, as discussed in this very recommendable tutorial, the unique identification of an H II region can be a bit of a challenge in some cases while being obvious in many others. If you discover some interest for this topic, you might really want to read it for a start:

The Qust to find H II Regions

This paper will allow you to do some respective practical exercises using the SDSS data base and tools right after reading...

Finally, here is a perfect case of a SDSS galaxy with bright blue H II regions (SDSS imaging).

Image

The proof that the H II suspicion is correct can be easily done with the SDSS tools, by selecting such bright blue regions and examining the corresponding spectrum. Here is it for the above example:

Image

You can clearly see the copious, characteristic H-alpha line at 6563 angstroem, which proofs the case.

Enjoy,
Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Thu, 13-09-12, 21:51 GMT 
Offline
User avatar

Joined: Fri, 29-02-08, 17:58 GMT
Posts: 85
This is getting better and better.

You know, that spectrum also shows a prominent oxygen peak in the green. When I look at the Tarantula Nebula in my telescope, I see it green. I have photographed it green too. Brighter nebulae such as M42 or Carina, instead, photograph red.

My camera is not an astronomy calibrated instrument, of course. I just found nice that fact that the two peaks are of similar intensity in this example.

G


Top
 Profile  
 
 Post subject:
PostPosted: Fri, 14-09-12, 9:31 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
abramson wrote:
This is getting better and better.

Thanks.
Quote:
You know, that spectrum also shows a prominent oxygen peak in the green. When I look at the Tarantula Nebula in my telescope, I see it green. I have photographed it green too. Brighter nebulae such as M42 or Carina, instead, photograph red.

My camera is not an astronomy calibrated instrument, of course. I just found nice that fact that the two peaks are of similar intensity in this example.

G


Right, the OIII (doublet) with nominal wavelengths of ( 4959, 5007) Angstrom is green in absense of redshifts. For galaxies with non-zero redshifts, the actual OIII color can vary substantially. In fact, there are examples of white, gray, brown, blue, purple, pink, green, yellow, orange and red OIII objects.


The transmission charateristics of the 5 standard (u', g', r', i', z') SDSS filters on the CCDs in the camera along with the mapping on blue, green, red screen colors:
Quoted from

  • u' = detector "sees" actual ultraviolet -----Screen display as nothing
  • g' = detector "sees" actual blue/green----Screen display as blue
  • r' = detector "sees" actual red---------------Screen display as green
  • i' = detector "sees" actual near infrared--Screen display as red
  • z' = detector "sees" actual infrared--------Screen display as nothing


(The imaging timing sequence is r' i' u' z' g').

Image

With this input, one may work out a table displaying the apparent OIII colors as function of the redshift z:
Code:
  z                  wavelength of OIII        actual colour of OIII      view on screen
==========================================================
0.00                      5007                          green                      blue
0.02                      5107                          green                      blue
0.04                      5207                          green                      blue
0.06                      5307                          green                      blue
0.08                      5407                          green                      blue
0.09                      5457                          yellow                     blue+green
0.10                      5508                          yellow                     blue+green
0.11                      5558                          yellow                     blue+green
0.12                      5608                          orange                    green
0.15                      5758                          orange                    green
0.17                      5858                          orange                    green
0.18                      5908                          orange                    green
0.20                      6008                          red                         green
0.22                      6108                          red                         green
0.24                      6209                          red                         green
0.26                      6309                          red                         green
0.28                      6409                          red                         green
0.30                      6509                          red                         green
0.32                      6609                          red                         green
------------------------------------------------------------------------
0.40                      7000                          nearIR                     red
0.60                      8000                          nearIR                     red
0.80                      9000                          IR                           not seen

At a wavelength of 7000 Ã…,  the OIII line will start to register on the IR channel and will be displayed as red on the screen. So expect to see some red OIII galaxies !!


Here is an excellent SDSS example of this striking color shift of OIII towards RED! According to the spectrum, it should be a dull grey, but the near IR is superimposed as red which gives it it's reddish colour in the SDSS profile(cf. above filter plot).

Image
You can easily see the considerable redshift of the OIII line (5007 Ã…) to the above 7000 Ã… from here:
Image

+++++++++++++++++++++++++++++
In conclusion: the paramount importance of a flexible COLOR visualization framework in the context of redshifted deep space objects can hardly be overemphasized!
+++++++++++++++++++++++++++++

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Fri, 14-09-12, 13:51 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
Meanwhile, I have also located the SDSS photograph of our reference dwarf irregular galaxy NGC 4449. It is most instructive to compare this SDSS profile based photo (left) with the corresponding one based on the Hubble profile (right)

Image

In this comparison image the rendering of the H II starforming regions in bright blue (SDSS) and pink (Hubble) is well illustrated.


Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Sun, 07-10-12, 12:46 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
I have just noted that I forgot to display some celestia.Sci pictures of Dwarf galaxies including the bright blue H II regions (as appearing with the SDSS color profile):

Remember, that the H_II regions are generated randomly according to a distribution that matches qualitatively the observed one! So there cannot be a 1-to-1 correspondence with photographs. I render irregular (dwarf) galaxies with so-called Perlin noise (for those knowing what this is ).

Here is a comparison of our reference dwarf galaxy NGC 4449 as rendered in celestia.Sci versus the same hires photo as above from SDSS:

Image

I have marked some conspicuous H II clouds in both images!

Next, here is a "medley" of various dwarf galaxies as rendered in celestia.Sci, with the blue background representing the pure H II color according to my SDSS profile. Since however various blue tones interfere, there is some amount of washing out.

[Click on image by all means!]
Image

This mixture of blue tones one also observes, when travelling to that dwarf galaxy shown in the top left corner of the image below. There I picked out a H II region illuminated by orange elliptical galaxies in the far background...

Image

Don't forget that with True Color RGB filtering these H II clouds would appear pink!

Enjoy,
Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Mon, 08-10-12, 8:57 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2726
Location: France, South, not far from Montpellier
Do you have control over the H II sprites independently from the rest? If so, perhaps a reduction of size of these sprites could be better... (they seems too diffuse within the whole structure compared to the ref image)


Top
 Profile  
 
 Post subject:
PostPosted: Mon, 08-10-12, 10:15 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4445
Location: Hamburg, Germany
ElChristou wrote:
Do you have control over the H II sprites independently from the rest? If so, perhaps a reduction of size of these sprites could be better... (they seems too diffuse within the whole structure compared to the ref image)


Since computer time for such critical rendering tasks is a crucial factor, for now, I did not use a special sprite (texture) for these H II clouds. I rather used a simple and fast criterion that selects roughly the correct points on the galaxies, which I then simply render with the standard sprites, yet colored in the H II color.

This procedure works very well for spiral galaxies. There the boundary of the selected H II regions is also sharper than in case of irregular galaxies (<=> Perlin noise).

In terms of dedicated H II sprites the cost would be a lot higher.
First one needs to code a dedicated sprite texture that has a smooth 2d random shape. The algorithm for doing that is standard in graphics coding. Something like that
is the result:

Image

These bright-blue patches then would be candidates for H II and would be placed approprriately.

Not sure whether it's worth the extra effort. I'll continue playing ...

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Mon, 08-10-12, 10:34 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2726
Location: France, South, not far from Montpellier
t00fri wrote:
ElChristou wrote:
Do you have control over the H II sprites independently from the rest? If so, perhaps a reduction of size of these sprites could be better... (they seems too diffuse within the whole structure compared to the ref image)


Since computer time for such critical rendering tasks is a crucial factor, for now, I did not use a special sprite (texture) for these H II clouds. I rather used a simple and fast criterion that selects roughly the correct points on the galaxies, which I then simply render with the standard sprites, yet colored in the H II color.


You know I'm not a coder, but if the H II coloring is attributed to some specific sprite, no way at the same time to say something like "hey you blue H II sprites, now you get this wonderful tone, you will also be reduced by two in size!" ... I do imagine it's not that simple, but I was thinking in a math attribute (scaling) inserted somewhere in the actual code rather than something new and more fps consuming...


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 16 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group