Info

206265 x ——-i——— = 78 arcseconds . (Equ. 5.6)

12x25.4x40

This field of view—117 by 78 arcseconds—would be a fine size for Jupiter and Saturn: either would float comfortably in the image. However, finding objects—even bright planets—with such a small field of view can be frustrating. In response, you will need to develop techniques for finding your target object efficiently.

One method is to insert a flip-mirror unit between the projection optics and the CCD camera. This will enable you to see a much larger field of view, and simplify locating and focusing planetary images. Another method is to attach a highpower auxiliary telescope in ring mounts so that you can align it accurately. Equip it with a reticle eyepiece that gives a magnification of 25 x to 40x. Once you have it aligned, the high-power finder technique can save you lots of frustration.

Of course, alignment itself is frustrating, because before you align the finder, you may not be able to find anything except the Moon. Work from low power visual to high-power CCD imaging. Start with an eyepiece at prime focus and align the finder on a bright star, then put the projection system in place and reacquire the same star with an eyepiece. Align the finder. This gets you close. Now put the camera on the projection system, find the star, focus, and center the star. Tweak the alignment of the finder until it is perfect.

5.8.3 Recording High-Resolution Images

The camera itself is third in the triad of high-resolution imaging factors. Because moments of good seeing occur unpredictably, an effective strategy is to take a large number of images and save the best. At an outstanding site, one image in every three may be worth saving; at a typical site, only one in ten might be worth keeping. Assuming that the primary optics and projection optics have been properly matched to the chip and pixel sizes of the sensor, pursuing the "many images" strategy means that the camera must not only be capable of making the relatively short integrations required for planetary imaging, but it should also allow the observer to make a large number of images rapidly.

Astronomical CCD Cameras. Few of these cameras are designed for imaging bright objects with short exposures. In general, those that can do a good job are equipped with a mechanical shutter that keeps the CCD in darkness except at the moment of exposure. Because speed and agility matter, CCDs with small pixel counts tend to be more effective for planetary imaging. It is no coincidence that many of Don Parker's spectacular planetary images were taken with a Spectra-source Lynxx camera (using the TC211 CCD with its 192 x 165-pixel array) equipped with a mechanical shutter.

Webcams. Unlike astronomical CCD cameras, webcams are designed for making lots and lots of images. Most have CCD or CMOS-based sensors with an integral Bayer array for color imaging. The shuttering is electronic, so there are no moving parts, and the sensor electronics are designed to produce up to 30 images per second at a resolution of 640 x 480 pixels (307,200 pixels). Although the Bayer array means that the effective resolution is about twice the size of the physical pixels, or about 12 microns, simultaneously recording three color channels confers a significant tactical advantage on the humble webcam.

Digital Cameras. For lunar imaging, digital cameras really shine. The Moon is large, bright, and loaded with fascinating high-contrast detail. Used afo-cally, it is easy to attain a pixel-matching image scale and still capture a large portion of the Moon's surface. Furthermore, digital cameras are designed for taking lots and lots of images. Used with eyepiece projection or a Barlow, digital SLRs are also great for lunar imaging. For planetary work, however, both types of digital cameras must capture several megapixels of black sky to get roughly 250,000 pixels worth of planet!

5.8.4 Making High-Resolution Images

Having determined your strategy with excellent telescope optics, a quality CCD camera, and enlarging optics that give you the desired field of view or optimum focal ratio for pixel-size matching, you must next develop a tactical approach to making good images.

Get good polar alignment. Although lunar and planetary integration times are short, you will need good polar alignment, or the drift in declination over the time you gather images will drive you nuts. An error of 1° in polar alignment makes Jupiter move its own diameter north or south every two minutes. Do you want to chase Jupiter, or make images of it?

Set the clock drive rate. For planets, set the drive to sidereal rate—but for the Moon, set the drive rate to "lunar" to compensate for the Moon's average west-to-east motion of 33 arc seconds per minute. Each time a planet drifts out of the field of view, you lose image-making opportunities.

Clean the projection optics. A converging cone of light at high focal ratio reveals even the smallest specks of dust as dark shadows. Clean the optics in your projection system, clean the window of your CCD camera, and take flat fields to remove the dust specks that escape your best efforts at cleaning.

Determine the integration time. You now need to determine the integration time. The goal is to set it so that the brightest pixels have values of at least half the full-well pixel value but less than two-thirds full-well. (For a 12-bit camera, shoot for top values between 2,000 and 2,800; for a 16-bit camera, between

Figure 5.16 The most important lesson to learn in lunar and planetary imaging is that raw images always look fuzzy. In this case study, the raw image of the lunar crater Copernicus appears soft and mushy. Note, however, that the sharp transitions from light to dark hint at detail revealed on the page opposite.

Figure 5.16 The most important lesson to learn in lunar and planetary imaging is that raw images always look fuzzy. In this case study, the raw image of the lunar crater Copernicus appears soft and mushy. Note, however, that the sharp transitions from light to dark hint at detail revealed on the page opposite.

Make a test integration of 100 milliseconds; then examine the histogram. If the highest pixel values are 1,200 (out of 4,096), for example, then double the integration time. If the integration time is considerably too long, the brightest areas of the image will be saturated. Any time the highest pixel values exceed three-fourths of full-well capacity, reduce the integration until the brightest pixels are around 2,500.

Remember that astronomical CCDs are extremely sensitive, so integration times may be surprisingly short. Even with long focal ratios, unfiltered images of Jupiter and Saturn may require integrations of only around 100 milliseconds (short enough to avoid severe atmospheric smearing); and even with green and blue filtration for color imaging, perhaps only 1 second.

Shoot lots of images. Save the good ones. Since you can't beat the seeing, join it. Take an image, make a snap judgement, and save it if it's good. Then take another. If you shoot ten to twelve per minute and save one or two, you're doing pretty well. Don't bother to spend a lot of time studying each one: the idea is to make snap judgements and take lots. In an hour of intensive imaging, you may save 60 to 80 images.

Expect mushy-looking images. Raw planetary images look terrible. There

Figure 5.17 After processing, hard-to-see low-contrast image details are revealed clearly.

However, compare the two images carefully, and you will see that every feature in the processed image was already present in the raw image. Image processing does not create detail, it only reveals existing detail.

Figure 5.17 After processing, hard-to-see low-contrast image details are revealed clearly.

However, compare the two images carefully, and you will see that every feature in the processed image was already present in the raw image. Image processing does not create detail, it only reveals existing detail.

is no other word for it. Even when everything is working right and one after another is popping onto the screen, raw Moon and planet images look mushy. Nine times out of ten, they come up wavered or smeared, and even the fabled "tenth image" looks pretty bad. Mushy images are completely normal. Save those that look a bit less mushy than the others.

Touch up focus every few minutes. Don't focus once and assume that it's right. Touch up the focus every few minutes. If you don't already have an electric focuser, you'll start to understand just how nice they can be. Electric focusing means no jiggle and shake when you focus, and a more repeatable movement from pushing a button. As soon as an image looks good, get back to taking pictures. The more pictures you take, the more good ones you'll get. After you have saved five or six images, go back to the camera's focus mode and touch up the focus.

Fussing with the focus may seem counterproductive, but if you touch it up frequently, you stand a better chance of hitting it right on the nose some of the time. If you actually get it perfect—and you'll know this because you'll start seeing a higher fraction of really sharp images—suspend the next touch up until the quality seems to degrade. The point is that with long focal ratios, in most telescopes the point of best focus will drift, and you will need to follow it by checking and correcting focus often.

Keep a written record. Keep a written record of the imaging session, especially key factors such as the start time and stop time of sequences of images, eyepiece or Barlow in use, the projection distance used, the exposure times, and any other information that you will need when you process and evaluate your image the next day. Note changes in seeing, when and how you shoot your darks and flats, and all crucial settings that you make. If you get great images, you'll need this information to repeat the performance.

Shoot dark frames. It is tempting to get lazy and skip dark frames. Don't do it. Since you are going to enhance your planetary images to bring out low-contrast detail, you need to make them linear and get rid of the hot pixels. Shooting dark frames is easy: at some point during the session, cap the telescope and shoot a dozen dark frames. They will be almost perfectly uniform since not much thermal noise builds up in a short exposure, but proper calibration definitely improves the quality of the lunar and planetary images. When you subtract dark frames, you're also removing the bias frame, which is necessary for flat-fielding.

Shoot flat fields. During the imaging session, shoot a dozen flats and flat darks. An illuminated flat-field box is the best way to shoot flats, because you can make the flats without pointing the telescope away from the planet. Determine the integration time just as you did for the planetary image: set the peak pixel value between half and two-thirds well. Save eight to ten integrations as well as an equal number of flat dark frames taken with the same integration time.

Flat fields are important because they remove dust and vignetting that otherwise plague planetary images. However, remember that a flat field is good for only one physical setup. If you change the projection magnification, remove and replace, or rotate the camera, or alter anything else that changes the path light takes through the optical system; you will need a new set of flats. The only change you should allow—an unavoidable one—is a small focus change. It pays to get everything set up and running right at the beginning of the night and avoid further alterations. Make the flats about two-thirds of the way through the imaging session.

The next day, select the best images. By the end of an intensive imaging session, you will be tired. You may have saved as few as 20 or as many as 400 images, and you have a dozen dark frames and a set of raw flats and flat darks for calibration. Shut down the equipment, pack up, and go to bed. Leave the processing (and the emotional highs and lows that will inevitably accompany the outcome of your efforts) for the next day.

Process the images when you have the time to do it right. Make the master darks and master flat, then process all of them exactly the same way.

• Tip: AIP4Win allows you to process a whole night's take of images using the Multi-Image Auto-Process tool. This takes the drudgery out of calibration. Work out a fairly aggressive standard enhancement, and then run that enhancement on every image. When you line them up and compare them, roughly one in ten will be noticeably sharper and less smeared than the others.

Processing the next day is a vital step in the "quality control" for your images. You need a night's sleep to review in your mind what you did well and what you did poorly during the session. If you process them immediately, because of the significant element of chance in lunar and planetary imaging you can easily (and mistakenly) credit poor technique with good results, or vice versa.

Archive your images. Archiving digital information has never been so cheap or easy. Save the best selected raw images, the master dark and master flat, and a processed version of the best images on CD-ROMs. If you begin to produce good-quality planetary images on a regular basis, submit them to an organization such as A.L.P.O. (American Lunar and Planetary Association) or the Planetary Section of the B.A.A. (British Astronomical Association), where they will become part of our permanent record of the planets' behavior.

5.8.5 Solar System Targets

The planets, the Moon, and the Sun differ in angular size, surface brightness, and spectral properties. It helps greatly to adapt your imaging techniques to accommodate the particular features of each object. For more information, and to participate in organized and directed imaging, contact organizations like A.L.P.O. and the Planetary Section of the B.A.A.

Mercury. Despite its high surface brightness, this fugitive planet is difficult to image because of its proximity to the Sun—but of course this only adds to the challenge! As an inferior planet, Mercury presents its full disk when it is far from Earth, and a narrow crescent when it is near. The best times for imaging are near quadrature, when the planet presents a half-illuminated disk 6 to 8 arcseconds in diameter.

To image Mercury, you can either catch it when it is high in the daytime sky, or try to image it in twilight through a long and turbulent atmospheric path. Of the two options, daytime imaging is probably best. Because its surface features— Moon-like maria and craters—are neutral in color, you can use a red filter to reduce daytime sky brightness. Its small angular diameter and high surface brightness call for long focal ratio.

Venus. Cloud-covered Venus has the highest surface brightness of any planet, making blooming and overexposure a significant problem for frame-transfer CCDs. The only features on its disk are cloud markings that have highest contrast in violet and ultraviolet light. Like Mercury, Venus is most readily imaged when it is near quadrature, and presents a half-illuminated disk 25 to 30 arcseconds in diameter.

To record cloud markings, you should make images through a deep blue or violet filter (such the Wratten #47) coupled with an auxiliary infrared blocking filter. For any planet but Venus working in the deep blue would pose a problem, but Venus is so bright you can expect reasonable exposure times even with a red-sensitive CCD camera.

Mars. You can see more features on Mars by webcam imaging than you can visually. This is because Mars has a high surface brightness, so you can use a long focal length with short exposures, and also because the surface features have their highest contrast in red light, where many digital sensors have their greatest sensitivity.

Mars comes to opposition about once every two years, then reaching an angular diameter between 14 and 25 arcseconds; however, significant detail can be captured any time that Mars is larger than 6 arcseconds, so the observing season is 4 to 6 months long. Mars presents its full face at opposition and shows a pronounced gibbous phase at quadrature. Because of the planet's rapid rotation, to make accurately registered tricolor images of Mars, it is necessary to obtain three good images within a 5-minute time span.

Its appearance varies enormously depending on the imaging wavelength. In blue light, the disk often appears featureless except for lighter clouds over the poles; but in red and infrared light, the dark surface features and polar caps stand out clearly. Mars observers are interested primarily in tracking clouds and dust storms and measuring the seasonal variation in the size of the polar caps.

Jupiter. With its large angular diameter, good surface brightness, and omnipresent and ever-changing cloud belts, Jupiter is the most rewarding of the planets to image. It ranges from 40 to 48 arcseconds in diameter at opposition, large enough to fill the frame of a TC211-based astronomical CCD camera with a 10-to 12-inch telescope.

Jupiter's clouds are fairly strongly colored; images taken through blue, green, red, and infrared filters bring out the differences clearly. The Great Red Spot is the best known colored feature on the planet, rivaled only by blue features that appear in the tropical and subtropical belts and zones.

Making tricolor images of Jupiter is quite difficult because of its 9.8-hour rotation period. To prevent misregistration, the three filtered images must be made within a 2-minute time span.

The Jovian Satellites. Jupiter's Galilean satellites show disks large enough to record and possibly show light poles and dark equators with telescopes of 12 inches aperture and larger. The largest satellite is Ganymede; this fifth-magnitude object is about 1 arcsecond in diameter at a favorable opposition. Its surface brightness is about 50% lower than Jupiter's.

To image Jovian satellites, use an exceptionally long focal length so that you are sampling the tiny images at twice the Nyquist criterion. Take a large number of images (a few hundred) to build a sufficiently high signal-to-noise ratio for aggressive image enhancement. Select the best 20 or 30 images, register them by centroid, and stack them. Process the resulting composite. For comparison, apply the same techniques to a fifth-magnitude double star such as s Lyrae, and a single star of fifth magnitude.

Figure 5.18 Opposite: Enlarging the image from a 10-inch f/9 SCT to match the 9-micron pixels of the detector allowed critical Nyquist sampling. This image captures diffraction-limited detail in the lunar craters Ptolemaeus, Alphonsus, and Arazachel. Image by Thierry Legault.

Saturn. Saturn's magnificent ring system inspires heroic efforts to make outstanding images, but it is a fairly difficult target because its surface brightness is about one-third that of Jupiter, and at 18 arcseconds the disk is considerably smaller. The ring system spans around 40 arcseconds. As exposures move from fractions of a second for Jupiter to full seconds for Saturn, seeing becomes a relatively more important factor in imaging the ringed planet.

Saturn's atmosphere is less colorful and has lower contrasts than Jupiter's. Every few decades, however, bright white clouds well up suddenly and then fade away over several months. In the ring system, the Cassini Division is about 0.4 arcseconds wide, and shows up as a dark band even when it is unresolved. The next time Saturn's rings are well presented, it would be interesting to attempt imaging the ring spokes discovered by the Voyager spacecraft.

Uranus and Neptune. These two gas giants are roughly 4 and 2 arcseconds diameter at opposition, and sixth and eighth magnitude, respectively. It is easier to image the disk of Uranus than Ganymede because the planet is much larger; but Neptune is only a little larger and considerably fainter. The multiple-image technique (described for Ganymede) should reveal their disks. Voyager showed Uranus as having a deep, featureless haze layer, but more recent Earth-based images have shown cloud belts. At present we see Uranus at an oblique angle. Neptune has both light and dark cloud features, but it is doubtful they could be imaged on the tiny disk.

Lunar Imaging. The lunar surface presents a myriad of wonderful details on a body with high surface brightness; it is an excellent target for high-resolution CCD imaging. Because the lunar surface has very little color, the primary uses for filters in lunar imaging are to remove residual chromatic aberration that might be present in a refracting objective or enlarging optics, or to reduce the light of this sometimes-too-bright subject.

However, different geological units on the lunar surface do have subtly different colors. It would be an interesting project to make matched images of Mare Imbrium, for example, through blue, green, red, and infrared filters, and combine these to make enhanced color images showing lava flows of different ages.

The surface brightness of the Moon varies with phase, and it varies across the illuminated disk. Its surface brightness is roughly 20 times greater at full than as a three-day-old crescent; and in the gibbous phase, the areas in full Sun with no shadows are about 10 times brighter than the shadow-rich regions under slanting illumination on the terminator.

One significant technical problem in lunar imaging is that the Moon has such a large angular diameter that stray Moonlight can scatter in the projection eyepiece and projection tubes, causing field flooding and other stray-light problems. A solution for this difficulty is to let through only a small section of the lunar image on the optical axis by placing a small mask made of brass shim stock at the focus of the projecting eyepiece. The opening in the mask should be large enough to illuminate the CCD fully, but small enough to prevent additional Moonlight from entering the optical system.

Solar Imaging. Imaging the Sun is similar to imaging the Moon, except that it is a million times brighter. To reduce the Sun's brightness, place a full-aperture aluminized glass or Mylar filter over the aperture. Standard filters made for visual solar observing are suitable. These have a neutral density between 4.5 and 5.0 (ND=4.5 to 5.0). Like the Moon, the solar surface is virtually colorless, so color filters are not needed.

Noteworthy solar features are sunspots, granulation, regions of plage (spidery bright regions) near the limb, and rare white-light solar flares. Because these features are dynamic on time-scales of a few hours, it would be worth the effort of making a few hundred images of a small sunspot in the course of a day and replaying the sequence as a movie to view the changes. CCD cameras can also be used with narrow-band Ha filters to capture images of prominences, monochromatic solar surface features, and solar flares.

Whenever you point a telescope at or near the Sun, you risk damaging it and/ or your eyesight. Pay close attention to precautions such as capping the finder telescope and making sure that solar filters are securely attached.

Was this article helpful?

0 0
Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment