Re: Astrophotography discussion forums?

On 2008-07-01 11:13:40 -0700, jcdill.lists@xxxxxxxxx said:

I need help with astrophotography software options. I'm using Keith's
Image Stacker on a Mac (MacBook Pro, OS X 10.5). I'm not an
astrophotographer - I'm helping a friend with his photos. I'm looking
for a forum where I can post questions about astrophotography software
- how to stack, how to subtract a dark frame, etc. The problem is
that most of the tutorials are about using the software for imaging
planets and that's not what I'm trying to do. I need to stack the
images to bring out the faintest stars possible, while minimizing
noise.

Thanks!

Here is a response.

Date: Wed, 02 Jul 2008 20:00:14 -0500
From: Rick Johnson <wa0cky@xxxxxxxx>
Newsgroups: alt.binaries.pictures.astro
Subject: Re: ASTRO: Image Stacker on a Mac?



thepixelfreak wrote:


Posting for someone who doesn't have access to binary groups. If you care you can respond to

jcdill.lists@xxxxxxxxx

I need help with astrophotography software options. I'm using Keith's
Image Stacker on a Mac (MacBook Pro, OS X 10.5). I'm not an
astrophotographer - I'm helping a friend with his photos. I'm looking
for a forum where I can post questions about astrophotography software
- how to stack, how to subtract a dark frame, etc. The problem is
that most of the tutorials are about using the software for imaging
planets and that's not what I'm trying to do. I need to stack the
images to bring out the faintest stars possible, while minimizing
noise.


Not being a Mac user I don't know what's available for it but the process is the same and a quick look at that software indicates its not what you want as it is designed for web cams. I don't know if it handles FIT images that a CCD takes. I assume we are talking CCD camera and FITs images here.

Most name brand cameras contain software that can do the initial calibration and stacking.

First you must have a series of darks. The number should be at least 10 or the number of image frames you are stacking, whichever is GREATER. While most software will scale darks for time and temperature I much prefer to use standard time and temperature for my images and thus don't scale the darks. If you must scale them you also need at least an equal number of bias frames as well. Bias frames are simply dark frames taken at the shortest exposure time your camera can take an image. Most imaging software has an option for taking such frames and handles exposure times based on its database of cameras. Usually "Bias" will be added to the file name automatically.

As I said, I don't use them since I take the darks at the same time and temperature I image at. I have a library, updated every couple months, of the temperatures in 5 degree steps. In summer I use -20C most of the time. Again the software should have a setting that takes these images with the shutter closed and labels them "Dark". Note some camera shutters aren't all that light tight so if taking them by day you should be sure the lens cap is on and there are no light leaks when taking Darks.

Next you need to take flats. This is more complicated. These are evenly lit images. There are several ways of doing this. I use T-Shirt Flats. For these you cover the scope's objective opening with enough layers of white material that when the camera is pointed to a part of the sky such that no sun hits the cloth, the exposure time is 3 or 4 seconds. You want this to eliminate shading from a slow shutter that plaques some cameras. You want an ADU count of about 30 to 50% of maximum. Adjust the material layers or exposure time until this is reached. Again, use at least 10 or the number of image frames, whichever is greater. You also need darks taken at the same exposure time as the flats.

Darks are stacked using median combine. This should be an option in any stacking software. Again, it often comes with the camera. This median combined image is your dark for calibrating you light frames. If you are scaling and use Bias frames they are stacked the same way.

Flats are also stacked using median combine then dark subtracted using median combined darks of matching time and temperature. Again, dark subtraction is an option in all CCD stacking software.

Now you have your master dark, bias if needed, and flat field. I guess I should mention that you MUST use the same binning mode for this as your images use. You can't scale up or down these to match an image taken with a different binning or even taken with a different camera of the same make and model at the same binning mode or taken on a different scope in the case of flats. For flats the focus isn't all that critical but all compressors, extenders etc. must be the same as used for the images to be calibrated.

Most image acquisition software will have an option to auto calibrate images as they are downloaded. That is they well subtract the master dark and apply the bias (if needed) and flat field automatically for you. Most save two versions, one calibrated and one not. Great lifeline if you told it the wrong calibration images! Something I've done more than once.

Once you have the calibrated image frames you will then stack them. If there are satellites or cosmic ray hits, planes etc. again use median combine or if your software supports it, a type of sigma reject routine IF you have at least 6 image frames. If less use average combine. If you are combining images taken at a different exposure time then I prefer to use add mode for combining. Note Average and Add combines give the best signal to noise ratio. You can't do better no matter what claims you hear. Next are types of Sigma combine. With more than 6 images they can virtually equal add and averaging. I don't use this unless I have some nasty transients to remove because it also removes asteroids I didn't, and now won't ever, know were there. The worst combine method is median combine in that it gives the worst signal to noise ratio but there are times when planes and satellites as well as the lack of having a sigma reject routine available make it necessary. All stacking software has Add, Average and Median combine. Better programs also have one or more types of sigma reject type routines.

Now this said, some newer chips from Kodak as well as much of the Sony line have very low dark current such that often you can get away with rather sloppy darks or even no dark at all in cooler weather. Not knowing what camera is being used I can't tell if that's the case here.

Also if you are using no filters and the CCD is rather small compared to the usable FOV of the scope, say an ST-7 on a 6" f/4 scope that I used to use. Flats may not be necessary for non critical imaging. Flats have several purposes. One is to calibrate for pixel to pixel sensitivity differences. This is necessary for photometric work for instance but today's cameras are so equal, except for hot and cold pixels, this doesn't really bother general imaging. Another purpose is to remove dust particles that cast donut shadows on the CCD. If there are no filters and the optical window is very clean these may not be a problem. The other main reason is to compensate for vignetting. This isn't often a problem if the FOV of the chip is small compared to that of the scope. There are other ways of correcting for vignetting though flat fields though flat fielding is the best.

Your friend really should get a book on CCD imaging to understand what it going on. They explain things in a lot more detail than I can here.
Many swear by Wodaski's book
http://www.newastro.com/newastro/default.asp
I found it mostly a manual for CCDSoft and Maxim DL. Since I use CCDSoft you'd think that fine but to me it WAS the manual and I already had that. In the end I learned little from it yet many love it.

I learned far more from the Berry Burnell Handbook of Image Processing
http://www.willbell.com/aip/index.htm
But it is far more expensive and comes with worthless software for a Mac user and to me nearly worthless for the Windows user as well as it is very outdated in today's world (calibrates and stacks however but doesn't support any sigma reject routines).

There are others out there but I haven't read them so can't say anything about them. Many websites have good tutorials as well. Google can turn them up.

There are some good free software programs for the PC but I don't know what's available free or otherwise for the Mac. Check what's available in your camera's software package. SBIG cameras always come with the needed software and many other camera come with a light version of Maxim DL that would do what you want. Maybe they assume PC only, I never paid attention.

Rick
--

thepixelfreak

.

Relevant Pages

  • Daily #4059
    ... Earth Flats ... Sky flats will be obtained by observing the bright Earth with the HRC ... biases and dark will be used to ... well as an entire program of cluster studies. ...
    (sci.astro.hubble)
  • Re: Dark indoor shots
    ... it hard to make the pictures in dark places look light. ... If you can't get enough additional illumination you must adjust the camera ... At the shutter speeds required for relatively dark rooms you ... But a mid weight tripod with a sholder strap ...
    (rec.photo.digital)
  • Re: What the F-stop is wrong with my Canon?
    ... camera SHOULD under-expose and render the white subject too dark. ... You need to shoot something that is neutral. ... Since green grass is very close to a neutral tone, ...
    (rec.photo.digital)
  • Re: Print stills question
    ... >> often come out FAR too dark. ... >> need of less exposure. ... or the length of time your camera will allow your shutter to ... I never meant to imply that the difference of lighting WITHIN a shot wasn't ...
    (rec.photo.digital)
  • Re: Flash for Canon 40D
    ... rooms I was taking pictures of was very dark even tho it had a window. ... the room was so dark that the camera would not take the picture with out flash. ...
    (rec.photo.digital.slr-systems)