Processing…

No-one is getting out much for astrophotography in Cumbria at the moment, as we are seeing endless cloudy skies.  Plenty of time to practise processing.

I made it up to my favourite dark lay-by near Tebay on 23 November and captured an hour of the Andromeda Galaxy M31.  29 frames of two minutes each.  If you’re wondering what a single two-minute frame looks like, here is one.  This is what you get if you open the shutter on your camera for two minutes.  The darkness of the background shows what an excellent location it is on the remote country road.  I’ve not done any processing on this frame, just converted it to jpeg for display here.

_DSC0020

300mm lens @ f/5.6 on Nikon D90 @ ISO 1600

Stacking the 29 frames and applying the usual processing tricks of adjusting the background, stretching the histogram, and a little colour saturation boost, gives this:

integration_DBE_Stretch

M31: 300mm @f/5,6, ISO 1600, 58 minutes
29 frames of 2 minutes.

That’s a shot I’ve wanted for a long time.  M31 nicely framed through a 300mm lens.  What more can we do with this?  Well, today’s lesson is all about High Dynamic Range “HDR” transforms: re-stretching the brightest parts of the image to bring out further detail.  Pixinsight, my chosen processing software, permits this:

integration_DBE_Stretch_HDR6

M31: 300mm @f/5,6, ISO 1600, 58 minutes
29 frames of 2 minutes
HDR transform

The effect of this adjustment is noticeable at the centre of the galaxy, where previously the detail had been hidden by the excessive brightness.

I’m also on a mission to use every scrap of data I can gather on M31, and Pixinsight also allows me to re-scale last year’s frames (taken through a 200mm lens at 5 minutes per exposure) and amalgamate them with the current set of 300mm frames.  Here’s the combined set, processed as normal:

200+300

M31: 300mm @f/5,6, ISO 1600, 58 minutes
29 frames of 2 minutes
Plus: 200mm @ f/f.6, ISO 200, 40 minutes
8 frames of 5 minutes

 

Next, the combined set with the HDR transform pushed hard to enhance the detail in the spiral arms of the galaxy:

Transformed

M31: 300mm @f/5,6, ISO 1600, 58 minutes
29 frames of 2 minutes
Plus: 200mm @ f/f.6, ISO 200, 40 minutes
8 frames of 5 minutes
HDR transform

 

I could be happy with that, but it’s quite hard on the eye, so I layer the original and the transformed versions together in Photoshop, which gives us the best of both worlds.

200+300layered

M31: 300mm @f/5,6, ISO 1600, 58 minutes
29 frames of 2 minutes
Plus: 200mm @ f/f.6, ISO 200, 40 minutes
8 frames of 5 minutes
HDR transform layered over original

There’s still a long way to go with this.  I’d like to take longer exposures without the stars trailing, and so get finer detail in the outer reaches of the galaxy’s arms.  Even at two minutes through the 300mm lens, the stars trail by a couple of pixels or so.  A couple of pixels doesn’t really matter for stars (they just appear slightly elongated rather than perfectly round) but it matters a great deal for the fine detail in the arms of the galaxy.

Stitching!

I now write a regular piece for the Eddington Astronomical Society, setting out what’s going to be on view in the night sky each month. In all honesty, it will probably lean towards astrophotography!  You can see the March 2014 version here.

One of the opportunities identified as coming up in March was the view of Jupiter reaching its highest point in the sky, elevated 59° in the early evening of the 13th. This looked like a challenging photo project, with bright Moon, bright Jupiter, bright and faint stars and the inevitable Morecambe Bay glow. It would also be at its best during twilight.

Sadly the Cumbrian rain was forecast for that evening, but we enjoyed a few days of high pressure just before then, with a couple of clear nights.

On Tuesday, 11th March, I strolled up to the Mushroom on Scout Scar with no more than camera and tripod. The 18-200mm zoom is pretty wide at 18mm, but I wanted a field of view of about 150°, so I set it to 24° and took nearly forty frames in a grid pattern looking south.

Photoshop can stitch shots together to make panoramas, but it needs clear points of reference in the shots and can only run on automatic. So, if it can’t identify the overlapping shots itself, there is no option for user-intervention. It simply doesn’t work on night sky mosaics.

That’s when I found PTGui, a programme that does exactly the same, but with a much wider range of projection options and, most importantly, the facility for the user to identify the overlap points (ie the stars) manually.

Here is the result, the view south from Scout Scar, Tuesday evening13th March. The sky isn’t flat, of course, and stitching frames requires distortion according to various rules depending on the projection settings. Circular projection seems to give the best compromise on distortion, then I cropped the result to a nice tidy rectangle.

All frames ISO 1600, 10 seconds 24mm f/5.6.

Big sky!  Nearly 40 frames stitched in PTGui, each 10 sec ISO 1600, 24mm f/5.6

Big sky! Nearly 40 frames each 10 sec ISO 1600, 24mm f/5.6, stitched in PTGui

PANSTARRS through floodlights in the sky

Inspired by several astrophotographers around the world posting images of the amazing tail developing on PANSTARRS, I decided to have another go at this comet.  There is a minor road / very long layby running parallel to the A591 on the hill between Kendal and Staveley, with unimpeded views to the north.  I reckoned it would be almost as good as driving into the Kentmere valley.

En route, I dropped by Helsington Church again, to see how the planetary alignment was changing over the Langdales.  Jupiter, Venus and Mercury were now in a straight line and all clearly visible.  I met a delightful family who had come to look at the sunset after spending the day in Kendal, and we passed such an enjoyable half hour with my binoculars spotting planets that I forgot to take photos.

On to the layby, and quite easy to set up with Polaris visible in almost enough twilight to see the mount dials.  Hold on though, that northern sky is getting brighter, not darker!

18mm f/3.5, ISO 1600 1 sec.

18mm f/3.5, ISO 1600 1 sec.

Yes the noctilucent cloud season is under way.  I can’t make up my mind about these clouds that are only visible at night – is this astronomy or meteorology?  Yes, they are really pretty to look at, with amazing variation and complexity in their structure, but that’s Cassiopeia up there, and my dark sky location is ruined.

18mm f/3.5, ISO 1600 1 sec. Two High Pass filters: 40 pixels and 4 pixels.

18mm f/3.5, ISO 1600 1 sec.
Two High Pass filters: 40 pixels and 4 pixels.

Sharpened in Photoshop, the structure resembles waves on a beach.

Prompted by a comment on Noctilucent Clouds 2013, I have experimented with sharpening these images using two Photoshop techniques: Unsharp Mask and High Pass Filter.

NLC plain

50mm f/8, ISO 1600, 3 sec.

The unsharpened image.

NLC USM 200% 4P

50mm f/8, ISO 1600, 3 sec.
Unsharp mask 200%, 4 pixels.

Unsharp mask – Photoshop’s main choice of sharpening tool, but the effect can be seen across the whole image.

NLC HP40SL HP4HL

50mm f/8, ISO 1600, 3 sec.
High pass filter 1: 40 pixels, soft light blend.
High pass filter 2: 4 pixels, hard light blend.

High pass filter – this sharpens only regions of higher contrast ie edges, that benefit from sharpening but leaves the rest of the image untouched.  I picked up this tip from Nik Szymanek in a workshop at the International Astronomy Show in Leamington.  Nik was using it to enhance detail in galaxy images, but the principle applies just as well to noctilucent clouds.  I prefer this sharpening effect which doesn’t introduce the “sharpened feel” artefact to the whole canvas.

Anyway, eventually the light fades and I catch seven reasonable frames of PANSTARRS.

PANSTARRS passing Mag 4.3 Yildun and its Mag 5.8 partner "24 Ursa Minor". Yildun is the next star to Polaris in the tail of Ursa Minor. 200mm f/5.6, ISO 3200 7x3min.

PANSTARRS passing Mag 4.3 Yildun and its Mag 5.8 partner “24 Ursa Minor”.
200mm f/5.6, ISO 3200 7x3min.

Yildun is the next star to Polaris in the tail of Ursa Minor.

North is to the lower left of the shot, and the glow from the pesky noctilucent clouds can be seen in the comet’s tail – which is pointing pretty much towards the Sun.  I shouldn’t really moan about the cloud, as the Sun is only 13 degrees below the horizon so it’s not exactly dark anyway (from an astronomical point of view).

PANSTARRS in Casseopeia

First attempt, a fairly optimistic shot over the garden fence on 19 April.

Casseopeia with PANSTARRS

18-200mm @ 65mm f/5.6, ISO 1600, 6 min
12 x 30 sec frames stacked in DSS

The clear sky had tempted me to set up the kit, but the urban glow catching the moisture in the atmosphere defeated any serious attempt to produce a reasonable photo. I thought that stacking several frames would help edit out the glow, but it doesn’t work that way.

Out to a new site on 20 April, in a layby part way up the Kentmere Valley.

C&P4x30

50mm f/2.8, ISO 800, 2 min
4 x 30 sec frames stacked in Photoshop

Skies were forecast to clear for an hour or so around midnight, but the waxing gibbous moon spread light throughout the sky. There were a few wisps of low cloud on the northern horizon. These 4 frames were stacked as layers in Photoshop.

My intention was to assess the suitability of the site for views to the north. Having escaped the street lights of Kendal, the next town north from Kentmere Valley is Penrith, some 19 miles away over mostly uninhabited countryside. It was pleasing to see there was almost no discernible urban glow in the northern sky.

C&P4x2

50mm f/4, ISO 800, 8 min
4 x 2 min frames stacked in Photoshop

Finally, four reasonable frames of 2 min each, relatively cloud free. Given that I was able to write notes by the moonlight, and the humidity was high enough for cloud inversion to begin settling in the valley (see bottom right corner of the photo), this is a pleasing result. Certainly I will visit this site again around the New Moon in about a fortnight. Cloud permitting…

M31 and PANSTARRS

When the Cumbrian skies cleared on 1 April we had the opportunity for several nights of catching PANSTARRS in glorious juxtaposition with the Andromeda Galaxy M31.

First up, some test frames on 1 April which turned into a nice capture that I’d had in mind for some time.

50mm f/5.6, ISO1600, 2min 4 x 30-sec frames stacked in DSS

50mm f/5.6, ISO1600, 2min
4 x 30-sec frames stacked in DSS

With a standard 50mm lens and some judicious cropping, this letterbox format shows Mirach (one of the guide stars used when finding M31: “from Mirach, hop right one star, then again, then down to the fuzzy blob”) with M33 faintly visible on the left, M31 on the right and PANSTARRS making its way in from the bottom.

Quite a windy evening, so I cranked up the ISO to 1600 and took just four good frames at 30 seconds.

On to the following night…

200mm f/5.6, ISO 1600, 15 min. 30 x 30 sec. frames stacked in DSS.

200mm f/5.6, ISO 1600, 15 min.
30 x 30 sec. frames stacked in DSS.

Quite breezy again, so I kept the exposures down at 30 seconds, but on the 200mm lens.  30 frames stacked for a 15 minute total exposure.

PANSTARRS is only about 7 degrees above the horizon here.  That presents a whole new set of problems.  At this angle, the line of sight goes through about ten times as much atmosphere as at zenith, multiplying the effect of water vapour on the incoming light.  From this location, Helsington Church, the view North-West also passes over the lights of Windermere and Ambleside, giving a street light skyglow to the bottom of the frame.  One of the beautiful features of PANSTARRS is its fan tail, glowing by reflected sunlight, so any attempt to process out the skyglow tends to eliminate the tail too.

300mm f/5.6, ISO 400, 40 min. 20 x 120 sec. frames stacked in DSS.

300mm f/5.6, ISO 400, 40 min.
20 x 120 sec. frames stacked in DSS.

Another shot I had framed in my mind, estimating there would be just enough room in the 300mm frame to put both M31 and PANSTARRS.  As the wind had dropped, I could get exposure up to 2 minutes, and grabbed 23 frames of which 20 were acceptable.

Finally on 3 April…

200mm f/6.3, ISO 200, 60 min 12 x 5 min. frames stacked in DSS

200mm f/6.3, ISO 200, 60 min
12 x 5 min. frames stacked in DSS

The birthday fairy brought me a polarscope this year, which dramatically reduces the time taken to polar-align the EQ3-2 equatorial mount.  In  a couple of minutes I can align more accurately than I used to get from 30 to 40 minutes of drift alignment using the camera.  Certainly it is good enough for 5 minute exposures at up to 300mm.  I must find time to test the alignment with longer focal length.  Anyway, this stack of 12 frames at 5 minutes each has cropped nicely.  M31 is not as clear as I would have liked (see M31: up to 200mm) but it was too low in the sky for that.

There was another very pleasing alignment on 5 April but the Cumbrian clouds had closed in.

PANSTARRS time-lapse video – processed

The unprocessed version of this video was rather disappointing. See PANSTARRS time-lapse video for details.

I took a fresh look at Fred Espinak’s videos of PANSTARRS, and compared our exposure settings. Fred’s videos used the same camera and lens as mine, but he chose f/5.6, ISO 800, 2 sec and ISO 1600, f/5.6, 4 sec. My settings of ISO 400, f/8, 2 sec are two stops and four stops respectively darker than Fred’s, and it shows.

This version of my same video has been reprocessed in Photoshop to boost the exposure by two stops. It’s crude, but shows how much better that choice of setting would have been. I have also taken the opportunity to sharpen the frames a little, and crop too. The cropping accentuates the judder of the movement caused by taking a two-second exposure every five seconds – and processing to play at 15 frames per second.

Fred Espinak’s video of PANSTARRS with the Moon is just stunning.  Arizona does have wonderful skies.  Cloudy skies in the North-West of England are very frustrating, but I live here by choice, so shouldn’t grumble.  On the other hand, it would be nice to have another go at this comet before it fades away.