Michael Purcell's Astrophotography
[ First Image
| Top
| Thumbnails
| MasterList
]
Driveway Astrophotography
(Notes on Purcell's Images)
This page is intended to describe how I take my images from my driveway. Fifteen years ago the area
was semi-rural, with reasonably dark skies that permitted casual observing of the Milky Way.
Today, it is completely suburbanized, and the Milky Way is just a memory.
The Telescope.
The CCD Cameras.
Preparation.
Imaging.
Processing.
References.
A pictorial view of these steps.
ST6 Log contains a NightTimes article from 1993 that
describes how I first learned to use the Santa Barbara Instruments Group ST-6 Imaging Camera. As mentioned
in the article, I completed my Messier list using a 4" Unitron refractor and a second-hand
homemade 8" Newtonian. It was the release of the ST-6 by SBIG that convinced me I could create
worthwhile astrophotographs from my home, even though I had given up on visual astronomy
under my increasingly light-polluted skies.
My first CCD images were taken through a used 10" Meade 2120 (f/6.3). In late 1994 I replaced the 2120
with a Meade 10" LX200 (f/6.3). Both scopes were purchased from Shutan Camera & Video in Chicago.
With the 2120 I had to mount the telescope on the equatorial wedge & tripod
every time I wanted to observe or photograph. The LX200, with its heavier and more stable fork mount,
is simply too heavy for this, so I bought a tripod dolly from JMI. Now I keep the LX200 fully assembled
at all times, and simply wheel it out from my garage. Not as good as being permanently mounted, but much
better than before!
I set up the telescope at the point on my driveway such that Polaris just clears the trees to the North of
my back yard. This same spot also has houses to the East and West, and more trees to the South.
(After light pollution, my second biggest problem is having to set up in "small sky country".)
| |
ST-6 |
ST-7 |
| Width (Pixels) |
375 |
765 |
| Height (Pixels) |
242 |
510 |
| Pixel Size |
23 x 27 u |
9 x 9 u |
Field of
View |
18.5 x 14 arcmin |
14.8 x 9.9 arcmin |
| A/D |
16 bits |
16 bits |
| Tracking |
Track and
Accumulate |
Dual CCD
Self Guiding |
1. It was a dark and stormy night...
A dark image is simply an image taken with the shutter closed. Dark images are used to capture
the effects of the CCD's heat and background electronic
noise. It is necessary for the dark image to be the same temperature and duration as the image.
Since it does not matter when the dark image is taken, ideal times are on cloudy nights. You can
build a library of dark images at a variety of times and temperatures. This avoids having to also take
a 30 minute dark image just prior to the actual 30 minute image.
2. Flat field images
Since all optics and CCD chips have various defects, it is necessary to be able to remove these
imperfections from your images. The flat field image is created by taking a picture of a neutral
gray card, or even the twilight sky (my preferred technique). Focus, temperature, and duration
do not matter for the flat field image. The ideal flat image should have some pixels that have reached
about 50% saturation. Theoretically, you should take a new flat field image every
time you attach your CCD camera to the telescope.
3. Selecting
I have a spreadsheet of all Messier objects, including most of the new Caldwell Catalog (see Sky &
Telescope, December, 1995, page 40-41). Given the
area of sky that will be visible overhead, I review this list along with the corresponding star charts.
4. Finding
Previously, I would use Epoch2000 to print a one degree square star field map using the Hubble Guide
Star Catalog. I would then take a test picture and compare it to the printed star field, and adjust the
telescope accordingly (assuming I had gotten close enough in the first place to find my location
in the star field). In early 1995 I upgraded the software in the LX200 to include the High
Precision Pointing option. As a result, the telescope usually gets the object in the CCD chip on
the first try. Then it is just a matter of centering the object.
5. Imaging
The ST-7 is a dual CCD camera, one for imaging and a smaller one for automatic tracking. I first
run a calibration process for the tracking CCD, then pick a guide star. Usually there is something
usable in the tracking CCD, but not always. If there is nothing to be seen, I reposition the object
off-center in the imaging CCD until a suitable guide star shows up in the tracking CCD. I then start
the picture. Depending on the dimness of the object, I will take a picture of 5, 15, or 30 minutes.
If the tracking fails at any time during the imaging the ST-7 will save the image . This happens if I have not
correctly calibrated the tracking, or if the guide star is lost due to brief cloudiness. Finally, I take a
quick look at the raw image and save it to the PC's harddisk.
6. Dark field subtraction
Using the ST-7 CCDOPS software I simply select a dark frame that has the same duration and temperature
as the image, and then run the Dark Subtract utility function against the raw image.
7. Flat field process
I also use CCDOPS for Flat Fielding the image. I have a few flat field images that I use regularly. In a few
cases there really is no need for this step, simply because the object is bright enough and the image
was fast enough. After this I run filters that remove "cool" and "warm" pixels. (Warm pixels are usually
the result of cosmic rays). After all this, the image is ready to be "sharpened".
8. Image processing
All image sharpening is performed with MaxIm DL (previously called Hidden Image). MaxIm DL implements an algorithm called Maximum
Entropy Deconvolution. This is similar to the process used to sharpen images from the Hubble Space Telescope
prior to correction of its optics. This step may take hours (or even days) before
I am happy with the new image. The goal is to convert stars from fuzzy, out-of-focus blobs, into
bright points, all without over-curdling any nebulosity in the image, or making it all look over processed.
Although I create the best focus I can when capturing the image, it is normal for atmospheric turbulence
to blur (convolve) the image. Also, as the temperature drops during the evening, the focus of the telescope
will change due to contraction of its metal frame.
9. Palette adjustment
I use MIRA A/P to read the 32 bit FTS file that was output from MaxIm DL. I experiment with various
palette utlities (histogram equalization, etc.) until I have an image that brings out as much
detail in the image as possible. I then save it as a RGB TIFF and convert it to a BMP. Finally, I load the bitmap into
the Windows image viewer LviewPro to create the full-size, half-size, and thumbnail JPEG images that I later
upload to my Internet Web homepage.
Click here for a pictorial view of these steps.
- MIRA image processing software
- Axiom Research
- ST-6 / ST-7 / ST-7E : Santa Barbara Instruments Group (SBIG)
- P.O. Box 50437
1482 East Valley Road #33
Santa Barbara, CA 93150
(805) 969-4069
- MaxIm DL :Diffraction Limited
- Cyanogen Productions, Inc,
25 Conover St.
Nepean, Ontario, Canada, K2G 4C3
Phone: (613) 225-2732, Fax: (613) 225-9688
cyanogen@cyanogen.on.ca
- An excellent Meade distributor: Shutan Camera
- 100 N. Fairway, Vernon Hills, IL
(847) 367-4600
- Epoch 2000 : Far Point Software
- Far Point Software appears to be out of business, having sold their product to Meade.
- LX200 : Meade Instruments Corp.
- 16542 Millikan Avenue
Irvine, CA 92714-5032
(714) 756-2291
- Jim's Mobile Inc. (JMI)
- 810 Quail St. Unit E
Lakewood, CO 80215
(303) 233-5359
- LviewPro : MMedia Research
- Attn: Leonardo Haddad Loureiro
1501 East Hallandale Beach Boulevard, #254
Hallandale, Florida 33009
email to mmedia@world.std.com, mentioning LView Pro in the subject line.
Send a fax to 1-305-458-9698
LviewPro
[ First Image
| Top
| Thumbnails
| MasterList
]
This page last updated on May 20, 2001
Contact Michael Purcell