Simple Tripod Digital Astrophotography


Starting out with simple compact camera and tripod astrophotography, I think the place to begin is that of wide-field portraits of your favourite constellations. You might want to try some shots at an "easy" subject like the moon, but that will normally be a disappointing start, because you will need an effective zoom-factor of around 40x or more to get a decent portrait of the man in the moon. (But you can take very nice pictures of moon-lit landscapes with just camera and tripod).

But let's go for some large constellations first. Just set your exposure time as long and your aperture as large (smallest f-number) as possible and shoot a score of pictures or so (yes, it IS nice that digital images are so cheap, but remember to bring a spare, fully charged battery along). Remember to shoot a couple of dark-frames also, i.e.: Cover the lens and shoot a few blank picture at the same exposure settings.

Now you will be eager to see, what has been recorded. Play-back on your camera LCD screen will show a few of the brighter stars at most - if anything at all. So, you will have to load the pictures to your PC before you can assess the results of your efforts. And here then is what you will typically see as one of the better pictures on an average day under less than perfectly black skies:

Recommended 1: Before you do anything else, make sure that you have saved your original file safely in an archive for possible later re-use and that you will be working on a copy. The reason is that the adjustments you make will be irreversible after you have saved them. Another reason is that after you have subtracted dark-frames and or added multiple files, you will loose your original picture's EXIF-data (i.e.: Information embedded by the camera in you image file on exposure data, camera settings, original date of recording etc.)

Recommended 2: If you are working on a high-resolution / high quality image where small details are important you may also consider working on a (large, uncompressed)) TIFF-file because if you save (as you will during the process) in the common, compressed and more manageable JPEG-format you will degrade picture quality somewhat each time you save your work. All detail will be preserved, when you save in TIFF-format. You may also save in windows bitmap, BMP-format but you should be aware that any EXIF-data will not be saved in that format.


A. Orion. 8 sec. exposure with Olympus C50 Zoom at 7.8 mm f/2.8 (click to see large image)

Huh, is that all? Well, in fact it IS all and that is more that meets the eye at first glance. Here is, what you have actually recorded with that tiny lens in just 8 seconds:


B. Same picture after a bit of enhancement in PhotoImpact12 (click to see large image)

What happened between A and B? Fortunately, our software can "see" subtle differences in luminosity that the eye cannot catch directly, so we have to amplify those differences to a level, where we can see them. There are more ways to do so, each producing slightly different results, but here is the route followed for the picture above:

a) Dark frame subtraction

First, subtract your dark-frame (click on a frame to see large picture): In Photoimpact, open both frames; go to ADJUST; then CALCULATION and finally find SUBTRACTION. Place the photo as foreground picture and the  dark-frame as the base-image and carry out the subtraction. (Similar functions are found in all digital imaging software of  better class, although the function names and details  may differ).

A.  (as A above) (click to see large image)

C. Dark-frame click to see large image)

D. Dark frame subtracted from A in Photoimpact

(click to see large image)

Lets delve just a little on the subject of dark-frame subtraction and removal of noise. If you compare A. and D. in large view, you will readily notice that a few - mostly bluish - "stars" have disappeared. These are "hot pixels" present both in A. and C. In addition there is some inevitable thermal noise that you cannot remove completely. To see this let's use the AUTO-LEVEL function in PhotoImpact. (Otherwise mostly useless or rather detrimental in processing of astrophotos). Here is the result:

A.2 (A above after AUTO-LEVEL)

(click to see large image)

C.2 (C above after AUTO-LEVEL)

(click to see large image)

D.2 (D above after AUTO-LEVEL)

(click to see large image)

The excessive noise in the upper left corner is a peculiarity in low-light situations for my Olympus C50Z that I don't see with my other cameras, (but they have their peculiarities as well!). Hence, I have subtracted the dark-frame twice.

b) Enhancement, using LEVELS or CURVES
Now comes the interesting part - to bring out the hidden stars in image D above. This is not done by merely increasing the overall brightness of the picture but rather by selectively increasing the luminosity of the brighter pixels over those of the darker ones. This can be achieved by either (or both of) the LEVELS or the CURVES functions (as they are called in PhotoImpact - other software brands may use different names, but the functions are basically the same). There is no single "correct solution" as to how and how much. In this respect, amateur digital astrophotography is as much an art as an exact science, but once you get the grip and feel of it, you will establish certain standardized steps and may achieve fairly consistent results. Anyway, here's how we got from A over D to B above, using LEVELS:


E. Image B above subject to LEVEL adjustment in PhotoImpact

F. Same image after adjustment of levels

You will notice that there are three sliders for "input levels". The level ("brightness") of a pixel (red, green or blue) will have a value between 0 = black and 255 = the brightest you can have. If you slide the white towards the left, you will notice that the number starting at 255 decreases while the brighter parts of the image (: the stars) become more intense. Likewise, if you slide the black one towards the right, the number starting at 0 will increase and pixels with levels beyond the threshold will blacken out. The grey slider will adjust the contrast or gamma ("hardness") of the tonal range producing a darker and "harder" image towards the right (small numbers) and a brighter and more "washed out" image when moved towards the left. This way, we can increase the brightness of our stars while subduing background brightness and low-light noise. There are more useful features and options that we shall revert to elsewhere.
c) Cropping and blurring

You will notice that digital stellar images differ from those captured on analogue film. Pixels on a CCD are obviously rectangular boxes and tend to produce "edgy" stellar images. Further, due to the very nature of silver halide chemistry plus the spreading of light in the emulsion layers, stars on analogue film tend to be more rounded and disk-like. If you would like to soften your edges, you can simply de-sharpen your final image using the SHARPEN and/or BLUR functions of your software. Also, you may want to CROP your final image just to show the "interesting" parts. Here's an example of that:

G. Cropped part of B above

(click to see large image)


H. Stellar images in G. softened by means of FOCUS

(that can also un-focus) in PhotoImpact)

(click to see large image)


One final note regarding image H above: When you de-sharpen the stars, you smear out their light and you will have to brighten the stars once again using LEVELS or CURVES.
d) Comparison with star map
Finally, it could be interesting to see, what have we actually got? Thus I have superimposed H above on a star map produced in "Cartes du Ciel". This sandwich is produced by inserting H on the star chart using the OBJECT function and making it semi transparent using the OBJECT PROPERTIES function of PhotoImpact. The image has further been rotated using the TRANSFORM function. Surely, a good digital imaging program is indeed mandatory for the amateur astro photographer !

I. Image H above superimposed on a star chart from Cartes du Ciel

(click to see large image)


Comparison with the star chart reveal that we have captured stars down to approximately magnitude 6 - which is about what one may expect with wide field exposures under a not perfect black sky. If you shoot smaller star fields at longer focal lengths, (say 3x to 5x optical zoom) you may gain a couple of magnitudes as we shall see in a following chapter. (Try aiming at a bright star in your view finder and zoom in on that star. You will soon find out what I mean then).

Technical note 1: Visual and photographic magnitudes are not the same. In particular, the CCDs of ordinary compact  digital cameras have a quite different sensitivity for light different wavelengths compared to the sensitivity of the human eye. For example these CCDs are much more sensitive in the red parts of the spectrum, which you may convince yourself of by watching your activated TV remote control on your cameras LCD display: In most cases, you will clearly see the infrared light pulses from the remote control.

Technical note 2: When superimposing images and maps as above, it is a requirement that the projection method used for the map is the same as that used by the camera. In both cases we produce an image of the curved celestial sphere on a flat piece of paper / a flat CCD. The camera only knows one way of doing this (tangential projection) assuming ideal optics while there are different geometric approaches for map projections. Furthermore the centre of projection for the map and the centre of the photographic image (the direction in which your optical axis was pointing) must be the same (as expressed in celestial co-ordinates). I shall revert to the issue on angular distances and their measurements in a follwing, technical section.

Now, that was a lengthy introduction to the simplest of all subjects, but I hope I have given you some inspiration and guidelines to get the most out of your first digital images - and that I have shown you that the fascination does not stop at the capture and download of your images but that there are many hours afterwards of exciting exploration of your heavenly accomplishments.



Copyright 2008 - Steen G. Bruun