Imaging

A simple introduction to electronic imaging

Video cameras

I've only dipped my toe in the water with this but it was sufficient to indicate the potential of this approach to imaging. Some really good results are possible using nothing more than a hand-held camcorder, which after all contains a CCD chip just like the MX5's, ST6's etc of this world. But given the intended domestic use of the device means that only the bright planets, moon and sun are suitable targets. The image below, of the crater Clavius, was recorded using a JVC camcorder hand-held behind the eyepiece of a 10" telescope at 300x magnification.

In the simplest approach your taped images could be viewed by just playing back on a television. But much better results are possible if further image processing is carried out.

To do this the image must be captured onto a PC using a suitable device. I use a Snappy made by Play (available from PCWorld). Cheaper cards like WinTV could also be used but the results are not as good apparently. If you record over say a minute this will generate many frames which can then be viewed individually and the best selected for subsequent processing. The advantage here is that the recording may capture moments of good seeing when the image is steady and therefore well defined and these frames can be selected. As with CCD techniques multiple frames could be stacked to create the final product, and various operations such as unsharp masking used to improve the image further. The image of Clavius above was made from just a single frame, unsharp masked with contrast and brightness adjustment.

Having explored the basic technique in this way it's then possible to move on to slightly more advanced methods. Setting the camcorder on a tripod for steadier recording for example. Alternatively the camera can be mounted on the telescope allowing the use of a drive if available. Also try playing with any adjustable settings you may have on the camcorder. Some allow limited exposure control for example. My own JVC has a 'twilight setting' which in certain circumstances greatly improves the image quality. And most of all experiment with the image processing software itself (and you don't need any of the specialist packages, PaintshopPro does a good job for instance) in order to make the most of the image. But remember not to overdo it. You want to extract as much as possible from the image but at the same time it should not end up looking artificial.

Ultimately of course camcorder imaging makes the production of movies possible. The rotation of Jupiter, motion of sunspots (using a suitable filter) etc.

Small security-type (CCTV) video cameras are more versatile than camcorders. They tend to be designed for working in low-light conditions (preferably <1lux) and it is usually possible to remove the lens allowing prime focus imaging (a lot of camcorders do not permit this). In addition they tend to be small and light thus making it easier to mount them at the focuser of the telescope. Monochrome is generally and although I've never used one some pretty amazing results are possible. 

PC cameras

Not all PC cameras use CCD chips, instead the detector is a CMOS device which doesn't seem to be very suitable for astronomical imaging. However, the CCD versions can produce some very good results. The monochrome QuikCam is a very popular choice, although they do not appear to be very easy to get hold of these days as they have been superceded by colour versions, though these too are frequently used. QuickCams are relatively easy to modify for astronomical use (see QCUIAG). However, any CCD-based PC camera could be pressed into service. I picked up a Sony video conferencing camera and was amazed at the results obtained just on my first effort.

These images of Saturn and Jupiter were my first efforts with the Sony camera attached to an undriven 6" Newtonian via a 2X Barlow lens giving an effective focal ratio of f19. The seeing conditions were good although a near full moon was close by. The lower pair of images show the result of stacking several frames together (6 for Jupiter, 3 for Saturn), whereas the upper pair are just single frames. Both sets were otherwise processed similarly with Adobe Photo Deluxe 2.0.

The camera was quite primitively modified by just removing the standard lens and cobbling together an adaptor from some scrap plastic tubing to facilitate attachment to the telescopes focuser. The camera was then connected to the Snappy frame grabber on a laptop and out to a portable TV so that the image could be monitored. Getting the image of a planet on a tiny CCD chip with a manually operated telescope is a challenge to say the least, and when you do get it it drifts extremely quickly. In addition this method of working means that you are limited to setting the Snappy in 'moving scene' mode, whereas a motorised telescope would permit the selection of one of the 'still scene' modes which would produce better results. I hope to experiment with this as well as further modifications to the camera such as transferring to a new box with a T-adapter and maybe air-cooling.

For loads more ideas (not to mention experience) on these techniques check out the QuikCam and Unconventional Imaging Astronomy Group, QCUIAG.

For some great images, plus a step by step guide to simple image stacking check Jim Ferrairas' page.