|
![]() ![]()
Welcome, fans of "Astronomy Picture of the Day", and those of you who attended my presentation at the 199th Meeting of the American Astronomical Society! If you don't want to wade through all the boring stuff below, you can go right to the Image Gallery. I've been interested in astrophotography since that night some 35 years ago when I hand-held my Dad's Topcon SLR up to the eyepiece of my Tasco 4.5-inch reflector and snapped some pictures of the Moon. The fact that they turned out at all was nothing short of a miracle in my mind. I've kept up my interest in astrophotography over the years, accumulating a few items to help facilitate the process. However, I have also learned that a real commitment to the art involves a fairly large outlay of cash and time, neither of which are particularly abundant in my Northern Virginia lifestyle. I bought a Celestron-8 in 1980, and for several years took snaps of the Moon and planets, and occasionally piggybacked a 200mm telephoto lens for large-scale deep-sky shots. But as 1986 drew near I knew I'd need a bigger gun for Halley's Comet. Fortunately I happened upon a relatively good deal on a Celestron Comet Catcher, which was the equivalent of a 500mm f/3.6 telephoto lens. With this combination I was able to get a nice portfolio of comet and DSO images, and even managed to get a few published in Astronomy, Deep Sky, and Sky & Telescope. However, at this time, folks like Tony and Daphne Hallas were "raising the bar" to the point where my images were nothing if rank amateur. I haven't used the C-8/CC combo for several years now, and the old orange C-8 has been replaced by a pre-owned Celestar-8, which I find very convenient for those PTA-sponsored star parties at the local elementary school. However, my job here at the Naval Observatory has allowed me to use one of the Celestron 8-inch Schmidt Cameras bought for Halley in '86. This contraption, coupled with some of today's fast color emulsions, has resulted in some very nice portraits (IIMSSM) of Comet Hale-Bopp and a few selected objects like the Andromeda galaxy (see below). The Schmidt takes some fine photos, but it can be a cantankerous beast and not easy to use as a portable setup.
The idea of digital imaging has always grated on my "purist" chemical photo side, but over the past several years I have to admit that the quality of digital work, especially in lunar and planetary imaging, has really taken off. So it was only natural that I began to think of some way to get into this field. A quick scan of any Sky & Telescope issue these days will certainly when the appetite of a potential digital imager, but a check on the ads for the most popular CCD cameras will bring reality crashing down on anybody with a tight budget...until now. An article in the June 1998 issue of S&T caught my eye as I was leafing through some back issues. "QuickCam Astronomy" by John Buchanan showed some very impressive lunar and planetary images taken with a $50 Connectix black & white QuickCam, one of those little doohickeys that sends a picture of your room and everything in it out over the Web. A little judicious hacking (Warning: cracking the case voids the warranty!) and you had a CCD camera not all that different from Richard Berry's Cookbook CCD.
A trip to the local computer superstore for some CD-R's found me returning home with a Logitech QuickCam Express, a color USB web-cam with a 320 X 240 CMOS imager chip and software to control it. Buchanan hacked his QuickCam pretty extensively, removing the PCB from its cute little round case and re-mounting it in the top of an "Edge" shaving cream can. I didn't want to get too surgical, so I simply uscrewed the lens assembly from mine and hot-glued a slightly modified Kodak 35mm film can to the round case.
The Moon was peeking through sucker holes in the clouds, but despite this I was able to get a couple of shots. This is the first one:
The next night proved to be clearer, so I dragged out the Celestar-8 and fired off a few dozen shots of various parts of the Moon and the double star Albireo. By this time I'd found a very good web site, the QCUIAG (QuickCam and Unconventional Imaging Astronomy Group). This led to a small freeware application called AstroStack, which I used to combine several frames, the results of which can be seen in the next picture. This image shows a "raw" frame of the crater Gassendi (left), along with a processed image made from a stack of 6 frames processed with a sharpening mask for 3 iterations.
This, in turn, led to a mosaic of the southern lunar highlands. This mosaic is made from 3 images (6 frames each, processed as above with AstroStack) pieced together using the demo version of PaintShop Pro 7.
I also processed the Albireo images, the result of which is this picture.
A few nights later I tried the QuickCam on the USNO's 12-inch Clark refractor. Despite bad seeing, I managed to get this image of Rasalgethi (Alpha Herculis), a 4.7 arcsecond binary star system. This pic was made from a stack of 10 images that was first sharpened, then blurred to remove the sharp mask's noise.
A few days later I tried to get a picture of the big sunspot AR 9169, a huge blot on the Sun that (at the time) was the largest spot group seen on the Sun in 8 years. Between the very large image scale of the 12-inch telescope and the bad seeing at 2:00 pm, I found myself hand-holding the camera to the barrel of the 4.5-inch f/10 Clacy finderscope to make the picture below. I had to completely wrap the camera in duct tape to blot out the internal glare, but I managed to snag 7 frames, of which 4 were useful for the image stack. An unsharp mask pulled out quite a bit of detail from the murky sunspot, but left some processing artifacts in the surrounding photosphere.
Finally, I decided to try some preliminary planet images. The images below were taken from my front yard in Alexandria, VA at about 1:30 am EDT on October 1, 2000, after I'd hauled myself home, rather bleary-eyed, from the NOVA Star Party. They were made with the Celestar-8 (with an added declination motor) and a Tele-Vue 2.5X Barlow lens. The images are composites of about 20 frames stacked and processed with AstroStack and run through an unsharp mask and a sharpening algorithm.
I also attempted my first foray into a big lunar mosaic. The result may be seen below. Click on the small image to see the full-size version. The 320x240 CMOS chip in the QuickCam Express lacks low light sensitivity for possible deep-sky imagery, but the Philips Vesta has a much more sensitive 640x480 CCD with 5.6-micron pixels. Some of the planetary results with this camera are pretty awesome. Sylvain Weiller, A French member of the QCUIAG group, has used this cam and a C-8 to get these images of Jupiter and Saturn.
Seeing these images gave me a goal to shoot for... In mid-October I purchased a Philips Vesta 675K web-cam and modified it as above (film can and hot glue), and the QuickCam Express was subsequently hacked back into an ordinary web-cam. "First light" on the Vesta was (auspiciously?) Friday the 13th. I'd set the C-8 up in the school playground near my house and fired off dozens of frames of Jupiter, Saturn, and the Moon. I was using another QCUIAG-developed program, Vega 1.2, to control the camera. Zip! Nada...nothing showed up on the bitmaps! That's what Friday the 13th will do to you! After midnight (that's when I finished reading the directions) I started in again. This time it worked. I stayed out until nearly 2 in the morning, then stayed up another hour processing the images. Frankly, I was stunned. I've never seen detail visually like that before through the Celestar! So here's what my first attempts with the new cam looked like:
Jupiter (with Ganymede), 10/14/00, 0506 UTC
and Saturn, 10/14/00, 0450 UTC Images acquired with Philips 675K web cam and Vega 1.2 software on Celestar-8 telescope with 2.5X TeleVue Barlow Lens. Shortly after I made these images, disaster nearly struck. I had decided to try to take some images with the USNO's 12-inch Alvan Clark refractor. The night was very clear, the air was stable, but it was cold for Washington. After attaching the camera, I accidently brushed it with my elbow. The hot-glue joint for the film can failed, and the camera fell about 5 feet to the floor. A more sound mechanical connection between camera body and film can adaptor was decidedly in order! However, I still wanted to see if the camera worked, so I duct-taped it back onto the telescope! These are the images made on that fateful might, after the "crash". Yes, folks, it's a robust little camera! I decided to use the screw-in lens assembly that came with the camera. Since I wasn't going to be using it as a web-cam anymore, I delicately drilled the plastic lenses out of the screw-in collar. I cut the top lip off another film can, and drilled a hole in the bottom (Kodak conveniently puts a small dimple there to mark the center) to pass the collar through. I then slipped a 1.25-inch washer over the collar (to add mechanical stability between the end of the film can and the body of the camera) and screwed this assembly back into the camera housing. The result looks like this:
On the left is a side view of the modified Philips Vesta 675K. On the right is an "end-on" view. Note the drilled-out lens collar screwed through the bottom of the film can. I have subsequently purchased a commercially-made adapter that accepts standard 1.25-inch screw-in filters. This unit cost about $20 and is available from Steve Mogg. This lathe-made adapter simplifies the centering of highly magnified targets such as planets and double stars and provides a very rigid support for the webcam. In the year that has elapsed since I started this project, the members of the QCUIAG have made great strides in both software and hardware. One of the members of the group, Steve Chambers, has found a relatively simple modification to the basic Philips Vesta and newer ToUcam webcams that allows exposures of up to 10 seconds. Early results with these cameras show promise for imaging deep-sky objects. A wonderful new French (with embedded English) image-capture software package, AstroSnap, is being developed by QCUIAG member Axel Canicio. This freeware program has many features, including focus control and image tracking, and it will control the Chambers-modified cameras as well as "native" ones. I have successfully used this control and capture software to image close binary stars and images of planetary satellites. I am also now using another freeware application, Registax, for image stacking and processing with very satisfactory results.
Right: Io & Europa, 2001 October 23, seperation about 3 as Both images made with 20-cm SCT, 2.5X Barlow Lens, and Steve Mogg Adapter Images captured with AstroSnap, processed with AstroStack and The GIMP The humble webcam will never replace the dedicated CCD astronomical camera. In terms of dynamic range, spectral sensitivity, and image scale, the CCD wins hands-down. However, the webcam brings the world of CCD imaging into the hands of anyone on a tight budget. It has great potential as a teaching tool for public school systems, and very satisfactory results may be obtained from urban areas for relatively little investment. Many of the processes used to enhance professional CCD images may be applied to webcam images, producing results that were beyond the reach of most astronomers 20 years ago. With the age of digital imagery now firmly entrenched, the potential of almost any digital imaging device is being rapidly explored. Digital cameras, webcams, and low-light video surveillance cameras are beginning to find interesting astronomical applications. The day will soon come when almost every amateur astronomer will have inexpensive digital imaging capability at his or her disposal. By Popular Demand...I'm working on a Gallery Page of work done with the Philips Vesta 675K WebCam. You may see this work-in-progress by clicking here. This page will be continually updated with new images as I get them. I hope you have enjoyed this little adventure in very modest digital astronomical imaging. Questions? Comments? Please e-mail me here.
Send comments on this page to webmaster@usno.navy.mil
|