Spiral arm alignment
-
Is the apparent different angle and distance from center of the clumpy spiral arm on the right from SDSS to DECaLS merely a processing / optical difference? ( can't overlay images)
Posted
-
by
JeanTate
in response to Ghost_Sheep_SWR's comment.
I would guess yes.
If you have a suitable image processing package - DS9 would be high up on such a list, but its User Guide is less than stellar - you could overlay the images to get an idea of just how different the two datasets are (DR1 and DR12 work from the same data).
Posted
-
Thought so, it's just that it looks like it's aiming straight down in SDSS and makes an angle in DECaLS, although i know it takes millions or billions of years to notice movement at these scales.
Sounds like it's a lot simpler using Photoshop with layer transparency or something like it than DR9 8)
EDIT: jigsaw puzzling
Looks ok now
Posted
-
by
JeanTate
in response to Ghost_Sheep_SWR's comment.
Sounds like it's a lot simpler using Photoshop with layer transparency or something like it than DR9
Yes! 😃
The major downside is that you can work with only the 256 intensity levels (8 bits) in the JPEGs, not the 12 (14?) bit ones in the original data ...
Posted
-
Doesn't really matter that much when comparing positions though.
Thanks!
Posted
-
by
JeanTate
in response to Ghost_Sheep_SWR's comment.
True.
One thing I forgot to mention: in most galaxies, the range of surface brightness (at the scale/resolution of SDSS and other ground-based instruments) is quite small. So small in fact that, with a careful choice of a stretch function, a JPG can comfortably display almost all meaningful surface brightness variations. And that's what the processing which produces SDSS and DECaLS (and more) images does.
So?
Well, in many, perhaps even most, such images, a decent image processing package (GIMP, Photoshop, etc) used on the JPG image can be used to bring out almost all the relevant features! 😃 Perhaps the easiest to reveal are low surface-brightness features (though careful treatment of the background may be necessary); next might be the subtle features which an unsharp mask can reveal.
Where might this approach not work?
Galaxies with bright nuclei, esp where there's an AGN (the range is too great); some clusters/HII regions in nearby galaxies (ditto); where the processing settings which produced the original SDSS (etc) image do not align well with the range of surface brightness (this happens, for example, near bright stars, or where the cutout image combines two tiles which have very different backgrounds).
Posted
-
That's informative, i've also noticed that even the JPG images usually contain some more information than at first can be seen by eye.
For instance, with only the standard built-in picture modification features on iPhone it's possible to enhance these different features. Especially with saturation, brightness and contrast levels, although from a certain point it starts to magnify background noise.
If i remember correctly i even managed to make certain features visible near a bright core, a couple of other iPhone modified examples can be seen in this (slow-growing, but interesting 😃 ) thread;
Posted
