Recent lunar work- Dec & Jan.
Graeme Coates
Hi all,
Typically, the only clear nights we’ve had up our way recently (and there haven’t been many…!!) have been when there’s a fairly strong moon, so I’ve been working on my lunar imaging a bit for the last couple of lunations using the 14”. (The scope still hasn’t been altered to get rid of the tube flexture, but I decided to rebuild the whole thing as a truss, and the first central brace is in production by a welder friend of mine who is handy with a MIG, so I can soon start to make the secondary cage.)
Two sets of images to post here – all use a ZWO ASI120MM camera - the first is at native focal length (1580mm) and is unfiltered. Images of the Straight Wall, Copernicus, South pole area (Tycho, Clavius, etc) and Mare Imbrium. Subsequent to this, I’d tried at higher focal lengths using a 2x Meade barlow – so running at about f/l=3.2m. I’d tried this unfiltered, and couldn’t get a decent result – lots of “double images” even in the stacked images. So, I tried using a filter and went with an Astronomik Planet Pro 742 filter – this is a near IR pass filter (cuts off virtually everything shorter than 742nm) and gives a result that is much less prone to atmospheric disturbance.
The results are in the second page below – despite the moon being 98% full (so little to choose from!) I’m pleased with how these came out (especially Pythagoras). I think with some decent seeing, I could improve on these and possibly try and even higher focal length. Of course, we’d need some clear nights with no gales to try again…
https://www.chromosphere.co.uk/2015/12/25/lunar-imaging-20th-dec-2015/
https://www.chromosphere.co.uk/2016/01/23/high-resolution-near-ir-lunar-imaging/
Cheers
Graeme
J R
--
You received this message because you are subscribed to the Google Groups "Altair_B" group.
To unsubscribe from this group and stop receiving emails from it, send an email to altair_b+u...@googlegroups.com.
To post to this group, send email to alta...@googlegroups.com.
Visit this group at https://groups.google.com/group/altair_b.
For more options, visit https://groups.google.com/d/optout.
Tony Angel
Tim Coskun
Graeme
These are fantastic images – so sharp. Forgive me if it already mentions this on your website, but did you have to do much in the way of processing the final stacked images in photoshop or the like?
Cheers
Tim C
From: alta...@googlegroups.com [mailto:alta...@googlegroups.com] On Behalf Of Graeme Coates
Sent: 01 February 2016 14:46
To: alta...@googlegroups.com
Subject: {Altair_B - 5493} -: Recent lunar work- Dec & Jan.
Hi all,
--
John Murrell
Hello Graeme,
I must agree with Tony your comet images are great and the Lunar ones pretty good as well.
The methods of processing images promoted by Hubl are interesting, particularly in view of the recent controversy on APOD and with the ‘winner’ of a Nikon photographic images submitting composite images to show something that was almost impossible to take. In the case of the APOD image it was the ISS occulting Saturn – The person who submitted the image said it was imaged from near his home. It did not take long for people to point out it was not possible to image the ISS and Saturn in daylight – the exposure for the ISS was too short to show Saturn. The photographer admitted it was a composite and APOD withdrew the page. The other example is on the BBC website where an aircraft was imaged passing the end of a stairwell which won a Nikon Amateur Photographer competition until it was pointed out the image was a composite.
That leads to the question is the Hubl method a composite and is at acceptable to present a view of a comet that is impossible to obtain ? My view is yes it is acceptable – people understand that the image has been heavily processed and the end result is far more attractive than showing multiple stars in different colours with a stationary comet. I wonder if APOD will accept this kind of Comet image in future after the controversy – I think they have in the past.
Regards
John
Graeme Coates
Thanks all. So the processing goes something along the following route:
1. Take video from camera using FireCepture – it helps to have a lot of RAM to setup a RAM disc and an SSD to get the highest frame rate. At 1280x960 resolution I can get about 21 frames/sec. Exposures are usually of the order of 2msec – be guided by the histogram for exposure and gain. Depending on the seeing, you’ll want lots of frames (eg 3000-4000). (NB: on the planets, esp Jupiter, it needs to be shorter as they rotate and blur!)
2. Import the video into Autostakkert and use the following settings:
· Surface stabilisation; Quality Estimator=Gradient; Noise (depends – try 3 as a start); Local (AP) as quality estimation method.
Set the stabilisation anchor to a nice big feature (crater!)
3. Click Analyse (get cup of tea!)
4. One you’ve set the quality, go to the image window and click place AP grid. You’ll need to play with the AP sizing – 50 seems a good start. Min bright at about 20 seems OK.
5. Click analyse again (this does the multipoint alignment). (More tea!)
6. Once complete, under Stack Option, set the frame percentages to stack – for a good number of frames, I’d try 2,4,7,15 as starts. Choose TIF as format. It’ll output the best x% of the frames based on these figures. You can set the drizzle option as well – I haven’t got on so well with this as yet – I struggle later with this…
7. Now open the TIF in Registax6 (try them all and see which is best). Then it’s a case of doing processing using the wavelets function – there’s lots of guides on how to do this, and it’s a bit of trial and error to see how the various options behave and what they do to the image.
8. Once you’re happy, save the TIF, open in Photoshop and then do a final sharpening and level adjustment.
Phew!
Here’s a reasonably decent guide to AS2 : http://www.astrokraai.nl/tut/guide_dennis_put.htm (though I use the Beta version)
Cheers
Graeme
Graeme Coates
Thanks both – the Hubl method was pretty hard to get a good result on. It took a long time fiddling around in Photoshop to get a result I was fairly happy with. Even then, there are a few artifacts in the tail if you look very carefully; the median combine doesn’t quite get rid of all the stars – especially the bright ones where it leaves trails on the fringes – you can just make these out in the tail.
I think the Hubl method is acceptable for this kind of thing though – it’s not a scientific grade image, and it shows the comet in a position relative to the background stars where it actually was in the run. The background and foreground are from the same raw data (and is much more aesthetically pleasing than the “RGB star blur” IMHO). You get more detail in the tail from a straight alignment based on the nucleus of course! The ISS/Saturn issue was that they were from two different runs, composited together, rather than from a real-time video of the transit. And the airplane/skyscraper one was just a complete mock-up.
I would have liked to have grabbed an image of Catalina, but the weather has not played ball – It was clear early last night, but Brize Norton (~5 miles away) was recording gusts of 50mph…
Tony, if you’ve got any good reduced data sets you wouldn’t mind sharing, I’ll have a go at a Hubl-type combine on them?
Cheers
Graeme
From: alta...@googlegroups.com [mailto:alta...@googlegroups.com] On Behalf Of John Murrell
Sent: 01 February 2016 20:51
To: alta...@googlegroups.com
Tim Coskun
I tried the Hubl method on Comet Lovejoy on 16 January 2015 and came up with the image attached. It was quite a challenge but given that I had not long started ‘serious’ astro imaging I was happy with the result. I got up at 4am one Saturday a few weeks ago to image Comet Catalina (there being a big ash tree in the way of imaging it earlier in the night) but have yet to finish processing the results as it is quite a time consuming method.
Best wishes
Tim C
J R
Sent from my iPad
<Comet Lovejoy 16 Jan 2015.jpg>
Graeme Coates
Nice! I suspect there might be more data lurking in there too – with the stacked comet image you might be able to pull up the tail a little more I think. The Hubl method has worked well though here :-)
GC
John Murrell
Tim & Graeme,
I found I could see a lot more detail in the tail by looking at my monitor at an oblique angle from above !
So the data is there – just not easily visible.
Regards
John