Jupiter 9/22/10 12:47am RRGB

Last night we had a nice break in the clouds, and along with unusually good seeing conditions for the Seattle area I was able to capture the following image of Jupiter.  This is one of our best Jupiter images to date.  Enjoy!

The above image was captured as follows:


2:30 of captures for each R, G, B

Stacked using Registax

RRGB image composited using WinJUPOS

Jupiter 7/15/10 4:50am RRGB

Here’s our latest Jupiter photo.  We had a great night of seeing and I was able to do a better job collimating the telescope.  More on this one later…

Jupiter 7/9/10 4:39am RRGB

Jupiter is currently rising and getting larger in the sky, and we should expect to see better and better images of Jupiter as the year progresses. In addition to the improvement in the altitude and visible diameter of the planet, I have also made some improvements to the technology we use to capture images.

The biggest new additions are the use of Point Spread Function (PSF) deconvolution in Fitswork, and compositing LRGB images in WinJupos using their cylinder mapping functionality.

PSF deconvolution allows me to reverse some of the imperfections in the imaging train. In this case, the scope was slightly out of collimation. I was able to reverse (some) of the collimation issues by using the shadow of Jupiter’s moon Io as the Point Spread Function for deconvolution.

The use of WinJupos for compositing images is a further enhancement. Using WinJupos I can shoot up to 2 minutes and 30 seconds of video for each R,G,B channel. Using WinJupos, each processed RGB channel is wrapped on a cylinder model of the planet, then rotated into the correct position for composition. This is a major technological advancement, as without the cylinder model I could previously only shoot about 40 seconds for each R,G,B channel. If I had filmed any longer than that, the planet’s rotation would have made it impossible to align the RGB channels into a single final image.

Here are the stats for this image:

Subject: Jupiter with shadow of Io
Date: July 9, 2010 4:39:18am PDT (11:39:18am UTC)
Location: Seattle, WA USA
Telescope: Celestron C11
Mount: Celestron CGE
Camera: Imaging Source DMK21AU04.AS
Filters: Astronimik Type IIc RGB
Imaging Train: C11 == NGFCM == Flip Mirror == Filter Wheel == Powermate 2.5x == DMK21AU04.AS
Recording exposure: 1/45 for all RGB
Recording framerate: 30 FPS
Recorded 4100 frames for each R,G,B
Stacked 800 frames of each R,G,B
PSF deconvolution in Fitswork
RRGB composition in WinJupos
Final processing in Photoshop

M27 6/7/10

Here is a photo of M27 (aka The Dumbbell Nebula) that we took just over a week ago from a dark sky site near Wenachee, Washington.  You can compare this against our earlier attempt at m27 from three years ago by clicking on the “Nebulae” tab and scrolling to the bottom of the page.  Clearly we have made some real improvements in our capture techniques, processing techniques and equipment.

We had some problems with passing clouds, so we weren’t able to integrate a whole lot of frames.  Total integration time was one hour and ten minutes.  Here are the details:

M27 – Planetary Nebula
10 exposures with 6-minute subs at ISO 1600
10 exposures with 1-minute subs at ISO 1600
Imaging scope:  Celestron C11 on CGE Mount
Filters/Reducers:  Celestron f/6.3 focal reducer (no filters used)
Imaging camera:  Canon XS/1000d with heatmirror replace with Baader filter glass
Guide scope:  Vixen ED80SF
Guide camera:  Imaging source DMK21AU04.AS
Captured with Nebulosity
Guided with PHD
Stacked in DSS
Post processed in Fitswork and Photoshop

M101 reworked

After spending so much time learning new imaging processing techniques on M88, I thought I’d apply some of those new techniques to the M101 image data from earlier in the year.

The results are what you see here.

For those who are interested, here is a copy of my image processing notes for Fitswork:

Load fits32rat of m101_9
Pixel Arithmetic -> Logarithmic
Use background flatten -> variable flatten
Blur Filter -> Median Blur 21,1
Color Image to B&W (Luminance)
Blur Filter -> Gaussian Blur Filter (Luminance) 10.4, 100
Pixel Arithmetic -> Multiply Value (Luminance) 0.05
(select color image)
Subtract Image (select gaussian blurred luminace)
Now take original Luminance image
Cutoff at high boundary 0.02
Cutoff at low boundary 0.02
for high boundary image, iter gauss sharp 4, 5, 400
for low boundary image, iter gauss sharp 2.4, 5, 400
Add low and high together
Image Combine -> L+RGB not scaled
Color -> Saturation

M88 5/7/10-5/8/10

Now that summer is approaching, we are starting to get more opportunities to do deep sky imaging.
The following was taken over two nights from Port Gamble, Washington – A fairly dark site that is near sea level.

M88 – Galaxy
30 exposures, with 6-minute subs at ISO 1600
Imaging scope:  Celestron C11 on CGE Mount
Filters/Reducers:  Baader UV/IR filter, Celestron f/6.3 focal reducer
Imaging camera:  Canon XS/1000d with heatmirror replace with Baader filter glass
Guide scope:  Vixen ED80SF
Guide camera:  Imaging source DMK21AU04.AS
Captured with Nebulosity
Guided with PHD
Stacked in DSS

Processing in Fitswork
* high/low cutoff to separate DSS stacked image into four layers
* iterative gaussian shapening on each layer
* wavelet NR on background layer
* add all layers together once sharpening completed
* star reduction
* rough histogram adjust

Color and exposure adjustments in Raw Therapee
* color boost/shift
* color NR
* exposure adjust, highlights adjust, midtones adjust

Final processing in Gimp
* final levels
* final crop
* star rounding (using Astronomy plugin)

I spent a good amount of time trying to figure out how to process this data.  Really an education for me and I have learned a tremendous amount about image processing from this project.

Note that there are several smaller galaxies (aka fuzzies!) present in the photo.  The irregularly-shaped object in the center right of the picture is Galaxy IC 3476.


M101 4/10/10 2:12am

Still very cloudy here in the great northwest.  We had a break in the clouds yesterday, and I took it as an opportunity to do some deep sky and planetary imaging from our home in Seattle, Washington.  Here are the results!

Subject:  M101 (aka The Pinwheel Galaxy)
Location: Seattle, Washington USA
Date and Time: April 10, 2010 2:12am
33 light frames @ 4 minute subs for 132 minute total exposure time
Modified Canon 1000d DSLR (heatmirror replaced with Baader filter glass)
Orion Skyglow Imaging Filter
Celestron C11 on CGE Mount
Guiding with PhD Guider using Vixen ED80SF and Imaging Source DMK21AU04.AS

Mars 2/11/10 9:09pm

There have been precious few opportunities for us to do astronomy here in Seattle for the past few months. A new baby in the family, and three months of almost constant overcast weather have impacted the amount of real telescope time we could put in. A few nights ago we had an unusual (and temporary) window in the clouds, so I took my C9.25 and shot a few sets of Mars video.

Seeing was average, with a persistent fast high-altitude atmospheric distortion. Transparency started out good but increasing got worse throughout the evening as clouds moved back in. Later in the evening this meant I did not have enough light at 1/60 second exposures and had to go to 1/30 second exposures. Unfortunately, the result was that the images later in the evening were not as sharp.

Notice the blue haze to the right and left of the planet surface. This is the evening haze on Mars, which is part of the normal weather pattern.

You may be wondering why this image is so much better than my previous efforts. Equipment and procedures were essentially the same. On 11/14 at 6pm Mars was only 8.8 degrees in size, and it was at 60 degrees above the horizon. The difference in size is due to Mars being close to opposition right now. Opposition essentially means that Mars is the closest to Earth that it gets for now. Weather also plays a factor and a night of steady seeing makes a big difference.

Location: Seattle, Washington USA
Date and Time: February 9, 2010 9:08pm (5:08 UTC)
Mars apparent diameter: 13.7”
Mars altitude above horizon: 53 degrees
3500 frames each R,G,B
Imaging Source DMK21AU04.AS @ 30fps with 1/60 sec exposures
Televue Powermate 2.5x
Astronomik Type IIc R,G,B filters

The following is one of other images of Mars I took later the same evening. As I mentioned earlier, the conditions degraded throughout the evening. This image was taken with the same setup as above, but with 1/30 sec exposures. You can see the effect of the slower exposures as the image is blurred along the direction of the wind (up/down relative to Mars) in the upper atmosphere.

Mars 11/14/09 6:00am

Here in Seattle we have had about four weeks of uninterrupted clouds. Last Saturday morning the Clear Sky Chart predicted clear skies around 4am PST – a perfect opportunity to image Mars.

Friday night I set up the scope, and Saturday morning at 3:50am I woke up to the sound of my alarm clock. Nice clear skies outside. I started setting up the scope and doing the collimation of the mirrors. Collimation is an important detail – especially for imaging planets. It needs to be dead-on or your images will turn out distorted. You do this by focusing the telescope on a single star in a section of sky near the planet you want to image. You then adjust the orientation of the secondary mirror on the telescope to cause the “airy disk” to appear perfectly concentric. I just about had collimation done and it clouded up again. Shoot!

I ended up doing work email for about an hour and a half while I waited for a clearing. I really wasn’t expecting it to clear, what with the rainy weather lately. I wasn’t ready to give up yet either, so I decided to hang out and see what the weather did. Luckily, around 5:45 PST the sky managed to clear up. Now with only 20 minutes before sunrise, I had a decision to make: Do I do finer collimation or do I shoot video? I decided to bag the detailed collimation and shoot the video.

What you see here is actually my first Mars photo. I’ve photographed Jupiter before, but never Mars. Note that the white area at the top of Mars is the polar ice cap.

Here are the image stats:

Scope: Celestron 9.25″ Schmidt-Cassegrain
Camera: Imaging Source DMK21AU04
Optics: Celestron 3x barlow, Astronomik type 2C R,B filters
3500 frames R 1/30 second exposure at 30fps, 1000 frames B 1/5 second exposure at 7.5 fps
RsGB technique

Image taken at 6am PST from Seattle, Washington USA


I used the “synthetic green” technique as described by Damien Peach in this article: http://www.damianpeach.com/marscolour.htm

Note that Mars was only around 8 arc seconds across (apparent diameter) at the time this photo was taken. To compare, the photos of Jupiter that I took earlier were around 45 arc seconds in apparent diameter. On the plus side, Mars is about 61 degrees above the horizon vs 25 degrees for Jupiter. A higher altitude in the sky means less atmosphere to look through, and the result is a clearer image.

NASA LCROSS Impact Site: Pre-Impact Image

Here’s a photograph I recently took of the expected impact site of NASA’s LCROSS (Lunar CRater Observation and Sensing Satellite). The satellite is expected to impact the surface of the moon at around 4:30am this Friday morning.

Link to NASA’s LCROSS Observation Campaign

Photo Details:

Camera: Imaging Source DMK21AU04.AS
Date Taken: 9/10/06 21:01 PDT
Filters Used: Astronomik type 2c red filter, 1/2 variable polarizing filter
Barlow: Televue Powermate 2.5
Telescope: Celestron 9.25 Schmidt Cassegrain
Exposure: 30fps with 1/60 exposures. 5000 frames

NASA LCROSS Cabeus impact site - pre-impact

NASA LCROSS Cabeus impact site - pre-impact

Should the weather here in Seattle be good (weather report not so promising), I plan to image the impact.