[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [APML] What's the Best Way to "Combine" Two Images ofDifferingTimes?
Hi Wade,
To integrate ("combine") two or more images with PixInsight, use the
PixelMath process. Here is the corresponding documentation chapter:
http://pixinsight.com/help/22_pixel_math/pixel_math_examples/pixel_math_examples.html
At the top of the page above is a description of some ways to integrate
images by averaging them. PixelMath works in PixInsight's native 32-bit
floating point format.
As Ray Gralak points out, integration methods based on statistical pixel
rejection (sigma, min-max, etc.) actually work when the set of images is
large enough - ten seems a reasonable minimum, based on practice.
If your images have different exposures, then you should apply correcting
factors when integrating them. Less-exposed images should be given lesser
statistical weights, since they have less signal and more noise. On the page
linked above you'll find also a description of how to apply such corrections
in PixelMath. Be aware, however, that film response is wildly nonlinear. In
practice, knowing the actual response curve of a particular film under some
particular conditions is not feasible. This means that you should derive the
correcting factors (weights) from actual measurements of pixel values. For
deep-sky images, I guess that the median values of the raw images could be
good starting values to establish relative weights between them, on a
per-channel basis. What do you think? Anyone has some thoughts with regard
to this?
Best,
Juan
______________________________________________________________________
Juan Conejero, Pleiades Astrophoto Team
PixInsight Home Page: http://pleiades-astrophoto.com/pixinsight/
----- Original Message -----
From: "Thomas W. Earle" <twade@bmi.net>
To: <astro-photo@seds.org>
Sent: Tuesday, September 20, 2005 6:23 AM
Subject: [APML] What's the Best Way to "Combine" Two Images of
DifferingTimes?
> For bright objects, such as M42, it is a known fact that combining
> several images at various times helps in getting the full dynamic
> range of this nebula; however, I'm not quite sure what's the best
> way to go about this. I have Registar, but none of the Functions to
> apply seem like the "perfect" choice. I feel using them will either
> destroy the shadows or the highlights. I see that many recommend
> using Ray Gralak's Sigma function; unfortunately, he recommends one
> should have ten or more images to combine for "best" results.
> There's no way I can get 10 shots since each film shot ranges from
> 30 minutes to 3 1/2 hours. Anybody have a proven technique to get
> the most out two or more images exposed differently? For example,
> let's say I take three images of M42 at F/5.0: one 30 minute, one 90
> minute, and one 180 minute. What's the best way to get the full
> dynamic range of these images (i.e. visible structure from the
> center of M42 to its faintest extensions? Any suggestions or
> comments would be greatly appreciated.
>
> Thanks in advance,
>
> Wade
_______________________________________________
Astro-Photo mailing list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo