Introduction 

I use a Philips ToUcam Pro for all my lunar and planetary imaging. Unlike deep-sky imaging, the targets are usually bright and often you need to use Barlow lenses to increase the effective magnification. The high magnification means you are also magnifying the turbulence in the atmosphere, which manifests itself as a shimmering image. However, the brightness of the targets helps because you can take very short exposures. If you take enough of these short exposures, a few will be taken during fleeting moments of stillness in the atmosphere.

This is why webcams are so good: you are able to take many frames (images) per second and save them in a “.avi” file. You can specify the frame resolution from some predefined ranges, up to 640x480 pixels. The ToUcam provides a simple utility called VRecord to control all of this this; if you do a lot of imaging, you may want something which streamlines the whole process, which is why I use K3CCDTools. This allows me to capture and save the avi files ready for processing later.

Once the avi files have been captured they can then be loaded into specialist processing software, such as Registax. All of us webcam imagers owe a huge debt of gratitude to Corr Berrevoets who has provided such a fantastic program for free!

Here are the steps in a little more detail:

Step 1 : Finding and Focusing your Target 

Let’s assume you have connected the webcam, it’s in the eyepiece tube of the scope and the image recording software (for me K3CCDTools) is recording what the webcam is ‘seeing’. However, finding and focusing your target is somewhat harder than you might think, for three possible reasons: 

Possible Problem 

Solution 

Firstly, at high magnification (strictly speaking, high f/ratios) you are looking at a tiny area of sky and if your scope doesn’t have sufficiently good pointing accuracy the target will be outside the field of view

My scope’s mount has the ability to move to a bright star close to the target of interest (called Hi-Precision mode). I actually prefer to use my planetarium software to select a nearby star. Once “synch’d” or “locked on” in this way, you can be more confident about its ability to point accurately when it slews to the target 

Secondly, you may not be perfectly focused, which may mean the target is there but too out-of-focus and blurred to see 

The solution to the first problem also helps with this problem: focusing on a planet is harder than focusing on a point source like a star nearby to the target. If you’ve just switched to a new focal ratio, say from native f/10 to f/20 with a Barlow lens, your image will be grossly out of focus. The star will most likely appear as a ‘donut’ shaped ring.

Use your focuser to coarsely change the focus until the ring shrinks to a rough point source.

For fine focusing, you may need to reduce the exposure (i.e. increase the frame rate) so the star is not washed out and use something like a Hartmann Mask to help you reach focus. An electric focuser helps enormously, avoiding the inevitable jitter which occurs when you touch the scope.

K3CCDTools also has a focus tool for fine focus which is very reliable.

You're now ready to slew to your target

Thirdly, the high magnification also means the light is spread out over a greater area and the target may be too faint 

You solve this once you’ve solved the previous two problems and have moved to your target. Here you adjust the exposure length (frame rate) until you can determine shading on the planet and no areas are over-saturated or too dim 

Step 2: Capturing the avi sequence

This is written from the point of view of a K3CCDTools user, but hopefully there will be similarities with other programmes. Step 1 should have left a nicely focused and centred image on the ‘live display’.

A few words on ‘exposure’, ‘gain’ and ‘frame rate’: the webcam is actually capturing a number of frames per second (the frame rate, fps), but each frame has a specific exposure, say 1/1000th of a second. You are aiming to capture images which are bright enough, but exposed with as short an exposure as possible, and to capture as many frames per second as possible. In practise, the brightness of the image will dictate how short an exposure you can use before the image is too dim and the inherent limitations of the ToUcam electronics will allow a maximum of 10 fps. Any higher, and the ToUcam automatically degrades the image quality to allow it to ‘keep up’ with the frame rate. (I believe there is a way to increase the frame rate and not suffer any image degradation by using a lower image resolution, say 320x240, but I’ve not tried that yet).You can also boost the ‘gain’ to make the image brighter, but too much gain can also increase noise. So you adjust both the exposure length and the gain to achieve a good bright image.

K3CCDTools allows you to specify how long a sequence to take and even a batch of sequences. For example, I could ask for 5 sequences to be captured, each 2000 frames long, repeating very 5 minutes. The output avi files are named automatically and it’s just so much easier than using VRecord.

So, how long do we need the sequences to be? Remember, we’re trying to get as many exposures as possible so we can extract the relatively few clear images which we can stack to get a good final image. However, there are two limiting factors on the sequence length:

·        Firstly, if the avi file is too large, your stacking software (e.g. Registax) may not be able to load it. I’m not sure there’s a hard and fast limit but I usually struggle with anything more than about 2400 frames at 640x480 resolution.

·        Secondly, if the sequence is collected over too long a period of time, the planet you are imaging may have rotated by a detectable amount over that period, effectively blurring the fine detail. This is a real problem for Jupiter, whose fast rotation means we have a time limit of around 2 minutes. For Mars, it’s a little more relaxed at 15 minutes and Saturn, around 5 minutes. Incidentally, the ToUcam is a colour webcam; for the more adventurous using monochrome cameras through filters, you have to image through each filter in turn within that time limit!

Anyway, K3CCDTools will automate the image collection process and you’ll end up with a batch of large (~1 GB) avi files ready for processing.

 Step 3: Processing the avi sequence

Here’s where the marvellous Registax is so useful. It is a powerful and easy-to-use tool for sifting through the avi sequence, extracting the best (sharpest) frames and then stacking those best frames. For the most part, Registax automates the process with default settings. However, there is an important job the user has to perform before letting Registax perform its magic: the user should scroll through the avi (using the slider at the bottom of the screen or the frame chooser) until a good clear frame is found - this will then be used by Registax as the reference frame, against which all other frames will be compared. 

Once the reference frame has been selected, you will need to put an alignment box around the target, or part of the target. For the planets, I select a box which covers the planet; for the moon I selected a smaller box around a bright crater. You can then press the Align button - if the Automatic check box was selected, Registax will proceed through each step of the process without intervention until you get to the famous 'Wavelets' window.

The wavelets sliders on this window are a way of selectively sharpening different type of feature buried in your stacked image. I'm sure there's plenty of science behind them but I tend to use trial-and-error until I'm happy with the result. There's a bunch of other things you can do on this final step, including alignment of the red, green and blue (RGB) colour channels. You will often see a blue hue to a planet, due to the different refraction the different wavelengths undergo - Registax has a stab at the alignment but I normally perform this step myself.

Before & After 

Not the best examples, but the two pictures here show a single frame from a sequence and a stacked and sharpened output. The seeing was atrocious, not really up to supporting f/30, but hopefully you can see the improvement. There are some better 'final' images in the Gallery section, but this shows a before and after 

Further reading

I have barely scratched the surface of this subject. I do not profess to be an expert and all I've tried to do is give an overview of the techniques I've used, which have been picked up over a few years. There are many online resources (such as Yahoo! groups), but I'd like to recommend a great book by Martin Mobberley: Lunar & Planetary Webcam's User Guide (ISBN 1-84628-197-0) as a fantastic introduction and more detailed insight into this area of astrophotography.


Make a Free Website with Yola.