My CCD Images.

M17 M42 M20

Images last updated 2000/04/07

A general note on how the images were taken. While CCDs are very sensitive and can take an image of very faint objects in a matter of seconds, this is not the way to get the best images. Almost all the images here are summed from several exposures each of several minutes duration, usually limited by bright stars in the field. Galaxy shots are typically 4 minutes, but objects closer to the Milky Way may be only 1 or 2 minutes. Summing lots of individual frames aids in beating the noise down. Unless otherwise noted, the images were taken through a 20cm f/4·5 Newtonian which yields a field of view of about 24·3 × 18·1 arcminutes. Pixels are 3·87 × 4·5 arcseconds, but have been resampled for these images so that they appear square. These images are JPEG compressed so some loss of detail has occurred. This is unavoiadable if I am to keep lots of images on-line. The TIFF files are somewhat larger.
More information on how I process images is here.

It is always difficult to keep the background level constant between various images (or at least I don't pay enough attention to this factor). If some images look a little dark, then please turn up the brightness on your monitor. The colour images require a 24-bit colour display to be seen to their best effect. If the images look poor, it may be that your browser has stolen too many colours from the look-up table; in which case save the image and view it later with some other viewer.

The images are presented in reverse chronological order - the oldest images are last, with the newest ones added at the top of the list. I think that my earliest images stand up well against many that I've seen, although the critical observer should note a general trend of improving image quality as time progresses.


NGC 2070
(42kB)
False colour This is a false colour image taken through an H-alpha filter (red channel), an [OIII] filter (green channel) and broad-band B (blue channel). Colour differences represent chemical differences in the nebula. The stars appear blue as they are dimmed through the narrow band interference filters and bright through the broader band of the B filter.
Mosaic of M42
(17kB)
H-alpha A 12-part mosaic to show most of the Orion nebula, shown at half scale. (Make sure you turn the brightness up to see all the detail!) Done through a H-alpha filter so that I could see the fainter parts without saturating the bright stars too much. 4 minute exposures, assembled with my mosaic software. The small stars and wide field give the image a more photographic quality, I think. Notice how the very bright star at the top was avoided. This (29Kb) image has been given the unsharp mask treatment! Dramatic, if unreal.
M83
(25kB)
tri-colour The spiral galaxy M83 is one of the best face-on galaxies visible in amateur instruments. This tri-colour image is the sum of 2, 4 minute exposures in R, 4, 4 minutes in V and 8, 4 minutes in B. It's been scaled for a more "visual" appearance showing detail almost all the way in to the nucleus. This method has the unfortunate side effect of significantly reducing contrast. (The green halos around the brighter stars were probably due to some passing thin cloud which appeared briefly during the session.)
Comet Soho (C/1998 J1)
(11kB)
tri-colour I must have got the colour right in this shot. Visually, observers couldn't agree whether it was green or blue - the same can be said for this image. The sum of 3, 30 second shots in R, 6 in V and 12 in B. Taken on the 29th May 1998 between about 08:20 and 08:40 UT.
Part of Vela SNR
(52kB)
H-alpha A narrow-band H-alpha image. This is just a small portion of the Vela SNR complex. This image covers about 20 arcminutes while the whole complex stretches for more than 3 degrees. Visually quite faint, but an [OIII] filter makes lots of it visible. A mosaic of this region would be great, but when I think of the number of images necessary...
Comet Hale-Bopp
(59kB)
It's still going strong. This is a mosaic of 3, 4 minute exposures taken on 1998/01/02. It covers about ¾° horizontally (declination) by 0·4° vertically. Note the strong spike in both directions. Visually through my 31cm the tail can be followed some 2°. This was my excuse to finally write some software for doing this - so expect some more mosaics to follow...
M42 !!
(22kB)
[OIII] Another image of M42! This was done with a Lumicon [OIII] filter plus V-band filter. Not as good as I had hoped, but interesting enough. The sum of 24, 20 second exposures.
NGC 2070
(28kB)
[OIII] At last a good image of the tarantula! This was done with a Lumicon [OIII] filter in front of the CCD. Because the filter has a significant red leak, a V filter was also used to isolate just the desired spectral region.
Horsehead
(22kB)
LRGB
tri-colour
I've re-processed my horsehead tri-colour with the LRGB technique to produce this image. The rotation isn't perfect between the unfiltered and colour frames, but it's good enough for the moment.
Helix
(29kB)
LRGB
tri-colour
The same colour image as below, but processed with a new (to me) technique. This involves keeping the colour information but replacing the spatial information with that from an ordinary, unfiltered exposure. The resulting image keeps the colour but improves the sharpness and noise characteristics. Details on this method are presented here.
Helix
(52kB)
tri-colour I've finally bagged the Helix in colour - the first time I've been out observing with the CCD for months! The sum of 3, 4 minute exposures for the red; 6×4 minutes for green; and 11×4 minutes for blue; making 80 minutes total exposure time. But worth it, I think. (Note the annoying satelite trail which went through one of the V exposures.)
Eta Carina
(41kB)
tri-colour Like my monochrome shot of the same region, it is the sum of lots of very short exposures in order to avoid saturating the bright stars. This time I used 20 exposures of 5, 10 and 20 seconds in red, green and blue. The nebulosity is consequently quite faint.
NGC 2070 (Tarantula)
(15kB)
tri-colour I'm a bit disappointed with this one. The dynamic range of this object is just too great to show both the central bright region and the much fainter nebulosity of the extended nebula. The green colour shows that this object - like M42 - is dominated by [OIII] emission in visual wavelengths.
M42
(yet again!)
(41kB) 
tri-colour Some of the fainter bits in tri-colour this time. Note how the 2 distinct colours delineate two regions of the nebula and shows the 3-dimensional nature of this object.
NGC 2024 (Flame)
(8kB)
tri-colour Near to very bright zeta Orionis and the horsehead is the flame nebula. Here is my tri-colour image. Scattering from zeta produces the blue halo on the left.
Horsehead
(28kB)
tri-colour The horsehead is quite a faint object to image in colour. Despite exposing for a long time, it is still faint. Note the asteroid image in the upper left - red, blue and green - from asteroid 6052 (Junichi).
Messier 42
(16kB)
tri-colour This is my proof that my tri-colour imaging technique produces colour that very closely matches what the eye would see could it receive enough light to fully activate its colour receptors. This is how I see M42 through a very large telescope. The green seems the right green and the red bar appears the right red. The image is made from 6 10 second exposures in R (red), 6 20 second exposures in V (green) and 6 40 second exposures in B (blue). A barlow was used to increase the image scale.
Messier 42
(16kB)
FALSE tri-colour This is what happens when you create a tri-colour image but use a different set of filters. This is using the V filter (green light) as the blue channel, the R filter (red light) as the green channel and I (infrared) as the red channel. Some people have suggested this as a way of overcoming the poor blue response of most amateur CCDs. If somebody can turn this image into anything even vaguely resembling reality I'd like to know how they do it!
Messier 42
(16kB)
FALSE tri-colour Still using the V, R and I filters, but this time using the I as the blue channel. There might be some sense in doing it this way as stars show up well in both the B and I bands. But as you can see, the green is totally wrong and the unnaturally blue stars just make it look... well, false. I'll stick to my attempts at true colours.
Messier 42
(15kB)
FALSE tri-colour Another false-colour image - but this time emphasising colour differences. This use U (ultra-violet) for the blue channel, V (green) for the green channel and I (infrared) for the red channel. This shows the hot, blue stars as blue (although saturated white in this view) and cool red stars (and dust-reddened ones) as red. The U-band only shows a few hot stars, while the I-band shows lots of stars.
Messier 33
(22kB)
The northerly declination of this splendid galaxy makes it difficult for me to do well. It is the sum of 8, 4 minute exposures done as the galaxy was well past it prime viewing position.
NGC 253
(22kB)
This is a splendid galaxy for both the visual and CCD observer. Its foreshortened spiral structure shows up well in this image, as do a few HII regions. It is the sum of 4, 4 minute exposures with the 20-cm Newtonian under adverse conditions - it was windy, cloudy and getting low in the west but it would be my last chance to get this for another 6 months and I missed it last time.
M16
(25kB)
I-band Where has all the nebulosity gone? This is an I-band image and at this wavelength the starlight penetrates the nebulosity far more effectively. You can still make out the familiar shape of the nebula but it is considerably fainter than in the optical wavebands. Like the H-alpha image below, it is the sum of 4, 4 minute exposures.
M16
(29kB)
H-alpha Another narrow-band H-alpha image. I've been wanting to do this object for ages but events conspired against me and so I've missed it for this season. When I did this shot it was very low to the horizon and so I'll probably try it again next time around.
The Helix
(40kB)
H-alpha The Helix through a H-alpha filter. The stars are tiny points but it doesn't seem to show any more detail than the unfiltered image below. This is the sum of 10, 4-minute exposures.
The Helix
(32kB)
Large objects like the helix are well suited to my system. I was a little disappointed that more detail wasn't visible.
M42
(37kB)
H-alpha A narrow-band image of M42 & M43 in the light of Hydrogen alpha. Using a narrow-band filter sepresses the bright stars and allows the nebulosity to come through. This image has been significantly processed as the dynamic range is very high. In particular, the unsharp mask filter in Photoshop was used. Compare it to this shot which is before the filter was applied. Another way to do it is like this which burns out the bright regions but allows the fainter outer regions to have a 'sharper' look.
M17
(21kB)
H-alpha A narrow-band image of M17 in the light of Hydrogen alpha. Shows detail not visible in the colour image below.
M17
(36kB)
tri-colour A tri-colour shot of M17. Lovely shades of red and a bright bar reminiscent of the bar in M42. Compare to the last image.
Comet Hale-Bopp
(45kB)
KG3 (minus IR) filter The comet on 1996/10/10 at about 10 hours UT. Taken through a KG3 filter, which effectively removes the infrared allowing only the blue-green-red light through. No processing other than dark and flat. The sum of 5, 60 second exposures.
Comet Hale-Bopp
(33kB)
The comet on 1996/08/10 at about 11:35 hours UT. It is unfiltered and is the sum of 4, 120 second exposures.
M8
(24kB)
H-alpha Taken two nights after the image below, but through an H-alpha interference filter. This one has been sharpened with an unsharp-mask filter.
M8
(34kB)
tri-colour A tri-colour image of the central region of M8. Note that it is not wholely red, but exhibits a fair amount of green. This is how I see it through a large telescope.
C1996/B2 (400kB) MPEG movie This is an MPEG movie made on the 19th March, showing 2 hours motion compressed into a quarter of a minute. The multiple-image mode of the Cookbook Camera software made this easy. I took a 15-second exposure every 30 seconds for 2 hours, making 240 frames in all.
C1996/B2
(87kB)
tri-colour Comet Hyakutake was a spectacular sight. I did very little CCD imaging of it, preferring to use conventional film. I did do this one tri-colour shot on March 15th.
NGC 2736
(112kB)
For southern hemisphere observers, the Vela supernova remnant (SNR) is a challenging visual object. This image is the sum of 15, 4 minute exposures and shows stars fainter than magnitude 21. Not bad for a 20cm homebuilt telescope and CCD. Compare it to this (64kB) section from the Digitised Sky Survey (DSS), which is from a plate taken at the UK 48-inch Schmidt Telescope, which goes to a limiting magnitude of about 22 in B (the original plate goes fainter, but the process of digitising and compressing the image looses a little). My 4 arcsecond pixels loose resolution compared to the photographic plate, but it shows just as many stars (but remember my image is an un-filtered CCD and red stars show much brighter than on the blue-filtered photographig plate).
Horsehead Nebula
(72kB)
Everybody has an image of the horsehead so why should I be an exception. Actually, I'm rather pleased with this shot. It is the sum of 5, 4 minute exposures and shows considerable detail within the "dark" nebula.
M42
(26kB)
H-alpha About the most famous object in the sky is M42, The Great Nebula in Orion. My image of the inner region was taken through an H-alpha filter to surpress the bright trapezium stars but still allow the nebula to shine through. Without the filter, the bright stars saturate in only a few seconds. The image was taken around the night of the full moon! Compare my image with the beautiful HST image of the same region (e.g. in S&T for March 1996, page 20) to see what extra is visible in a billion dollar telescope. To increase the image scale I used a Barlow lens on the 20cm telescope.
Eta Carinae
(49kB)
H-alpha The strange star Eta Carinae and some of the surrounding nebulosity are shown in this image, taken through a narrow-band H-alpha filter.
NGC 3372
(65kB)
The nebula in the region of the strange star Eta Carinae is both very large (some 2° across) and full of bright stars so is very difficult to image successfully with a small CCD. This small section is composed from 34, 15 second exposures. Even with such short exposures many stars are still saturated and produce ugly stripes, known as "bleeding". Note the satellite trail on the right.
73P
(23kB)
This is a later shot (1995/12/18) of Comet Schwashmann-Wachmann 3 as it was departing, showing a nice anti-tail.
73P and M62
(28kB)
(Schwassmann-Wachmann 3) I was very lucky to grab this shot as the comet moved past the globular cluster M62. Taken 1995/10/21 09:59:30 UT (mid-exposure).
Moon 6
(19kB)

Moon 8
(25kB)

Moon 20
(16kB)

I-band I've done very little imaging of the moon or planets because I don't really get the superb seeing that is necessary. I did one short session on the moon because I hadn't been able to go observing for over a month and the first available night had this big, bright thing in the sky. So I thought that I'd have a go at it. But my normal imaging telescope is not well suited at such imaging so I did it this way:-

I attached the CCD to my 31cm f/6·5 Dobsonian and placed a small Barlow lens in front of the CCD. This brought the image scale to about 1·5 arcseconds/pixel. I calculated that in an exposure of 50msec, the moon would trail about 0·75 arcseconds and so should not be readily detectable. However, there was still too much light so I added a 127mm (5-inch) off-axis mask over the telescope and an I-band filter. These 3 shots are some of the results. I think they're OK, given how they were taken.

R Corona Australis
(19kB)
The dusty region around this variable star is rather spectacular both visually and in this image, I think.
NGC 1313
(28kB)
This galaxy is on the edge of our local group. Easily visible (if you know where to look) on this image is the strange "supernova" SN1978K at around magnitude 19.
NGC 6822
(32kB)
Also known as Barnard's Galaxy it is in our local group of galaxies. The sum of 8, 5 minute exposures.
NGC 55
(23kB)
is another galaxy in Sculptor.
NGC 247
(22kB)
is one of the Sculptor group of galaxies.
NGC 1365
(32kB)
is a very nice barred spiral galaxy in Fornax.
M27
(41kB)
I point my telescope north of the equator occasionally. Here is M27 just to prove it.
M8 and M20
(27kB)
Here is a wide-angle view of the Triffid and Lagoon taken through my 171mm f/1·9 lens. Bright stars in the field make this a difficult region to image.
M20
(30kB)
Here is a what the Triffid looks like in monochrome. This image is the sum of 6, 1 minute exposures.
M20
(81kB)
tri-colour Here is my first ever tri-colour image. It is of M20 also called the Triffid Nebula. I have re-processed this image using my re-sampling technique. It is little different from the old image (28kB), although the background levels may have changed a little, and I've not painted over the pixel bleeds.
More LMC nebulosity
(33kB)
While I'm on the LMC, here is another small section showing some nice nebulosity and star clusters.
LMC nebulosity
(100kB)
Quite near to the Tarantula is this patch of nebulosity.
LMC
(74kB)
R-band
mosaic
I took this mosaic of the Large Magellanic Cloud through an f=171mm, f/1·9 lens. It is made up in 3 sections and each section is the sum of 9 images. An 'R' band photometric filter was used. The pixels are rather coarse with this lens, but it shows the galaxy rather well. The Tarantula Nebula is visible in the upper left side of the image.
NGC 2070
(26kB)
Also known as the Tarantula Nebula is in the LMC. It is the only image on this page (apart from the moon ones) that consists of just a single frame.
NGC 4945
(25kB)
A near edge-on spiral galaxy in Centaurus, only a few degrees away from NGC 5128.
NGC 104
(34kB)
The globular cluster 47 Tucanae was my third astronomical CCD image. The large dynamic range of this cluster has been compressed in this image making it appear somewhat "flatter" than in real life.
M 83
(30kB)
This face-on spiral in Hydra was the SECOND astronomical CCD image I took with the camera.
NGC 5128
(17kB)
Also known as the radio source Centaurus A, this famous galaxy is in Centaurus. This image was the FIRST astronomical CCD image I took (apart from focus frames).

homeback to Cookbook CCDto image processing


Page last updated 1999/04/07
Steven Lee