Author Topic: HDRComposition exposure question  (Read 2263 times)

Offline Ken Pendlebury

  • Newcomer
  • Posts: 47
    • View Profile
HDRComposition exposure question
« on: 2010 October 07 22:46:15 »
Hi folks,

I think that I would like to try the new HDRComposition on M31, but before I go out and collect data, I have a couple quick questions:

I haven't quite resolved the entire process in my brain and I am wondering if the decreasing exposure times need only apply for Lum or should this pattern be applied to RGB as well.  For example, can I take Lum at 10 min, 5min, 2 min and 30 sec and then a series of RGB all at 5 min?  Will the core be tamed by the Lum channel or will the color combination still blow it out if I don't take smaller increments of RGB (2 min & 30 sec)?

Thanks for any advice,
Ken
Regards,
Ken
My Astro Photo Stream
http://www.flickr.com/photos/kenastro/

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 6838
    • View Profile
    • http://pixinsight.com/
Re: HDRComposition exposure question
« Reply #1 on: 2010 October 08 01:16:59 »
Hi Ken,

Quote
I am wondering if the decreasing exposure times need only apply for Lum or should this pattern be applied to RGB as well.

You need a set of different exposures for both luminance and chrominance. Not only that: you need a set of coherent luminance/chrominance pairs.

I would shoot RGB exclusively. An HDR LRGB image can be quite problematic to assemble if you want to achieve a coherent chrominance for the whole image. Everything is much easier if you acquire just RGB.

To perform a color HDR composition, the HDRComposition tool must be used with RGB images. In this way a single scaling function will be computed for each frame, and the composition masks will be consistent for all colors. I wouldn't try to perform three or four HDR compositions (one for each individual RGB or LRGB component) and then create a final color composition.

Hope this helps.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/

Offline Ken Pendlebury

  • Newcomer
  • Posts: 47
    • View Profile
Re: HDRComposition exposure question
« Reply #2 on: 2010 October 08 01:35:18 »
Thanks Juan. I am pretty new to LRGB imaging. Isn't the luminance required for a certain level of detail?
Regards,
Ken
My Astro Photo Stream
http://www.flickr.com/photos/kenastro/

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 6838
    • View Profile
    • http://pixinsight.com/
Re: HDRComposition exposure question
« Reply #3 on: 2010 October 08 03:20:23 »
Hi Ken,

Quote
Isn't the luminance required for a certain level of detail?

Our vision system has evolved to perceive detail through the luminance, not through the chrominance. In this sense it is true that most of the detail that we perceive in an image comes from the luminance component. However, I see no reason for which we could achieve more detail by acquiring luminance and chrominance separately. The presence of small-scale structures (which support the high-frequency image components that we (humans) identify as 'detail') in the final image will depend on a variety of factors (seeing, focal length vs. pixel size, aperture, quality of optics, focusing accuracy, ...) which are all of them invariant to the fact that we acquire RGB or L+RGB.

It is true however that acquiring luminance separately you can achieve deeper images for a given exposure time, as a clear filter allows more light passing through, and hence you can accumulate more light in less time. In this sense, if time is a constraint then you can 'acquire more detail' with a separate luminance, so to say, as by going deeper you'll increase the SNR.

But it is also true that if you accumulate more luminance then the final image will lack chrominance, which has two consequences: some dim structures will have no chrominance support (they will be grayscale), and a low chrominance SNR will force you to stretch the chrominance to the point where chrominance noise can be objectionable. For this reason the chrominance is usually acquired in binning mode (2x2 binned pixels for example are four times more sensitive) at the cost of decreasing spatial resolution in the chrominance.

We have discussed the LRGB vs. RGB topic before on this forum. My personal opinion is that LRGB is only useful to save time, but the best results can only be achieved acquiring pure RGB with lots of exposure time. When time is the main constraint, LRGB with binned chrominance can be the only option to achieve the required depth. For HDR color images however, I still recommend you acquire only RGB, for the reasons I explained in my previous post.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/

Offline mmirot

  • PixInsight Padawan
  • ****
  • Posts: 879
    • View Profile
Re: HDRComposition exposure question
« Reply #4 on: 2010 October 08 09:28:52 »
I have been think this would be a good subject for article to submit to a major publication such as Sky & telescope.
Perhaps you should write this up along with a few screen shots of PI worked in.  ::)

Max

Offline Ken Pendlebury

  • Newcomer
  • Posts: 47
    • View Profile
Re: HDRComposition exposure question
« Reply #5 on: 2010 October 08 10:35:21 »
Juan,

Thank you so much for taking the time to explain that.  It helps a lot!!

Regards,
Ken
Regards,
Ken
My Astro Photo Stream
http://www.flickr.com/photos/kenastro/