Monday, April 21, 2014

Autostakkert / Avistack - a casual comparison

In this post I will compare the output from two of the best programs for sorting, aligning and stacking numerous exposures: Autostakkert v2.3 and Avistack v2.0. The subject is solar H-alpha images. The latest Avistack version is from July 2010 so it is not exactly undergoing a hectic development! Autostakkert is newer and is in a more dynamic development phase with minor updates occurring many times pr. year.

I did not spend a lot of time tweaking the processing parameters with Autostakkert - instead I deliberately did a 'quick run' using most settings at their default values. Avistack I know quite well and have been using for solar image processing for the past two years. During this time I have tweaked the settings to best suit my data. Hence, in a way the comparison is not 'fair' - that's why I called this a 'casual comparison'! See the cropped images below for the results, figure 1 is a prominence and figure 2 is an active region around a sunspot.

Figure 1 was made using 1500 frames from a Coronado filter on a 70mm f/7 refractor. The best 60 were used for the final image which then had wavelets applied to sharpen it (same settings for both images of course).

Figure 2 was made from 700 frames taken with a Daystar filter on a 6-inch f/6 refractor. Again, the 60 best frames were used but this time no wavelet processing was applied.

Fig. 1: comparison on a prominence.

Fig. 2: comparison on a sunspot.
The results are not dramatically different; both Autostakkert and Avistack do a very good job. However, the Autostakkert result is generally slightly better with just a little more 'omph' to it. Details of the settings are given at the end of this post - notice that Avistack has many more parameters to set. I have tried to choose settings for the programs that were similar to each other, but of course I cannot rule out that some Avistack settings exist that would allow matching or even surpassing Autostakkert. However, I do know Avistack quite well and have tweaked its parameters extensively in the past - I find that the results do not depend super critically on the chosen parameter settings.

In conclusion: Autostakkert gives a slightly better result and has much fewer parameters that require setting - a clear advantage, in my opinion. Some will balk at this 'lack of control' but sometimes there are just too many things to adjust and you can never be sure that the correct settings have been found. I find that Autostakkert is simple to use and works very, very well 'right of out the box'.


Some of the parameters used for the comparisons:

Autostakkert: image stabilization: surface, noise robust: 3, AP size 50, HQ refine: on, Drizzle: off, number of frames to stack: 60.

Avistack: quality analysis: on, frame cutoff: 60, area radius: 35, search radius: 28, alignment type: surface, ref. point min. distance: 13, structure threshold: 44, lower cutoff: 0.23, noise reduction: 0, quality area size: 32, ref. point alignment area radius: 24, ref. point alignment search radius: 2

Sunday, May 19, 2013

Throwing out the focuser

Every summer - where the Danish nights are too bright for deep sky imaging - I have fun doing Solar imaging instead. Being able to enjoy astronomy on a nice summer day without the cold and fatigue of a winter night is great!

I do high resolution imaging using a 6" f/6 achromatic refractor and a Daystar Quantum SE 0.5Å H-alpha filter. A Baader TZ4 tele-extender increases the magnification four-fold. When building such a setup the imaging train on the rear side of the telescope becomes really long and this leads to significant problems with flexure. Over the last few years I have optimized my setup to employ threaded connections everywhere, thus reducing flexure to a minimum. The imaging train currently looks like this:

Imaging train for high-res Solar imaging in H-alpha light (click to enlarge).
A special feature of this setup is that the telescope's original focuser has been replaced by a fixed length tube. Instead, focusing is done with a small helical focusser just before the camera. This way only the camera is moved and not the entire imaging train. This makes for much less flexure and focus slippage while shaving off nearly a kilogram from the total weight on the rear side of the scope.

Yesterday was my first time out this year with this setup and it was also first-light for the fixed-length telescope adapter. I had measured out very carefully how long it should be, but it was still scary-exciting to see whether or not I could get a sharp view using the little helical focusser at the camera. Fortunately, it worked just fine!

I only had a few minutes before having to go out but I managed to shoot active region 1748 which just a few days before unleashed four X-class flares in quick succession. More flares are likely but nothing happened during the two minutes I shot it.

AR1748, best 60 out of 2200 frames (click to enlarge).


Sunday, November 25, 2012

Eclipse 2012 - what went wrong?

It's now been eleven days since we stood in the shadow of the moon and I have now been back home in Denmark for a week. After recuperating and trying to remember what my day job is about I finally have found the energy for a very unpleasant task: finding out what went wrong on eclipse day.

Recall, that we got to see the second half of totality, with only a thin veil of high clouds present. Recall also that my ambitious project of high resolution automated imaging through a 4" refractor was a complete and utter failure.

I have worked on this project for several months. When I think about how close to success I was and revel in the details of how it failed I can almost feel physically sick. But that is exactly what I need to do; revel in details, go through it all again and try to learn from it so that I can do better next time.

Here's what happened. I set up OK, got the scope pointed at the partly eclipsed sun. Got the solar image focused and framed OK using liveview in Imagesplus. Got Eclipse Orchestrator started. Entered the precise geographic coordinates and local time from my GPS. Waited, thinking that all was ready.
Nervously awaiting totality with set-ups ready and too many clouds.

That was my second mistake (I'll get back to what the first mistake was later). I should have noticed that several small details were wrong. First, the displayed UTC time was one hour late. Second, the simulated view of the Sun did not match reality: the moon was leaving the solar disc instead of encroaching upon it.

When I discovered this we were only 15 minutes from totality and everything was just accelerating. I also had to function as the astronomical tour guide for our party of 28 paying guests. Frantically, I double checked the GPS coordinates and time, nothing was wrong. I wasted precious time restarting the PC and camera. With only seven minutes to go, I converted my script into an emergency version with only the coronal shots present (never practiced that). However, as totality began I could not start the script. Thinking that the system had somehow frozen up I realized all was lost and let it go. Instead I just enjoyed the rest of the show.

Yesterday, I went though it all again. Connected the camera, started Eclipse Orchestrator entered the local time on the PC as 6.36 AM, November 14, 2012 timezone GMT+9. Just as I had eleven days before. I saw the exact same problem - instead of totality being two minutes away the moon was just a few minutes from fourth contact! Fiddeling around I discovered, with a sickening, sinking feeling in my stomach the root cause: I had chosen GMT+9 as my timezone - it should have been GMT+10!!!!

Consequently, Eclipse Orchestrator thought that we were one hour further ahead in time than we actually were. How stupid of me.

Oh, well.

I must have been mixed up mentally by the pressure in the days since we arrived in Queensland. The tour program was very tight and I had been away from the equipment for three days because we had gone to the outback and the Daintree rainforest. I do remember entering GMT+9 on the laptop several days before the eclipse. That was a ticking time bomb and I could not overcome it on eclipse day due to the mental and temporal pressure of the unfolding events.

In the end of all this I arrive back to the feeling that all eclipse chasers have when totality ends, no matter what: when can I get to see this again?

Animated satellite images showing the lunar shadow racing across Australia and out into the Pacific (click to enlarge)

Tuesday, November 13, 2012

Eclipse 2012 - the event

We got to see the eclipse!!!

We left the hotel as planned and arrived on time. The site was open, not commercial'ish and we had a private area at the best spot, looking over the Captain Cook Highway and out over the bay. The area was warm, there was not a lot of wind and there were no bugs. The air was very moist and clouds formed and dispersed very actively. Sometimes we a got a very slight drizzle. Not the best situation, but we have known all along that this eclipse would occur under complex and active tropical conditions.

Setting up, then waiting.
I set up the equipment: Borg 4" refractor on astrotrac with Nikon D300 controlled by Eclipse Orchestrator. I also got the H-alpha scope up and running for the guests. Finally, I put an old analog Nikon camera with color film and a 55mm lens behind our crowd so that we would be framed with the eclipsed Sun. Quite a lot of stuff to tend to this time. I am pushing it to the limit - got to remember to enjoy the eclipse with my senses and be mentally - even spiritually - engaged.

 
Venus rising.
All was ready. Venus rose - bright and steady - where the Sun soon would follow. We all took that as a good omen.

After first contact just eight minutes after sunrise, things seemed to progress very rapidly. The time from first contact to totality spanned just 54 minutes. Measured GPS position and time; entered into the software. Found the sun, focused using ImagesPlus, then went back to Eclipse Orchestrator.

Fifteen minutes before totality things started to go wrong. Large clouds started drifting past the sun. I noticed that Eclipse Orchestrators simulated view of the eclipse did NOT match reality. I scrambled - was the GPS coordinates entered wrong (no!), was the time (no!). Reboot and try again, same problem. If the software is not in synch with reality the images will not fire off at the right moments and all would be lost. Just seven minutes to go. Instead of just dumping the project and dedicating myself to enjoy the spectacle I frantically made an emergency script, using the last few minutes with my head at the computer screen. Suddenly totality started, but a cloud was blocking the sun - none of the usual ahhs and oohhhs from us all; more like a tense mumbling.

One percent makes all the difference.
I tried to initiate the emergency script but found that Eclipse Orchestrator was not responding. The fully eclipsed sun came out from behind the cloud only covered by a thin veil of haze. VERY BEAUTIFUL! All was lost with my imaging project. I ran up to the analog camera and fired off some shots of the spectacle.

I think we got the last two thirds of totality including the diamond ring and Baily's Beads. I spent too much time fiddling around and the whole thing ended before I could find myself mentally participating. Too much ambition, too little time. Leaves me a bit frustrated.


The inner corona and numerous prominences.
Still, the luck we had in seeing the totality at all was amazing. Moments later a large cloud came by and blocked the sun for ten minutes.Although a bit frustrated I am mostly very happy. I had fun doing the preparations and I will probably find time to analyze what went wrong and try again some other time. For the eclipse chaser there always comes another time!

Be sure to click all the images to get a better sense of what happened this morning. The photographer is Inge-Lise Krylbo, one the participants on this journey. More photos will follow in the coming days and we'll also hear about how it went for other observers here in tropical Australia.

Eclipse 2012 - this is it!

It is 3.00 AM. Spent all night preparing. Sky looks promising; about 70% clear. Packing up and eating some food. Will meet with one of the guests who knows about weather and confirm that we are going to our primary site.

I am nervous about forgetting some essential piece of hardware so I better finished. We are leaving in 40 minutes. Next update will be after totality!

Eclipse 2012 - update before totality

>>>>>>>>  This update posted 1.5 hour before we leave the hotel  <<<<<<<

I put the gear together - went OK. Found that the 4" scope was out of collimation, as expected after the outback roads. Got it collimated OK. For this I used my artificial star, but had to pause halfway into the process because it started to rain. These tropical drizzles just come and go all the time! Scary.

Checked the Eclipse Orchestrator script once again. Found that the image sequences for the Baily's Beads and the Diamond ring effect were switched around on both sides of totality!! I got them switched around so that prior to totality the Baily's Beads comes first, then the diamond ring. Reverse after totality. Whew... Good I discovered that! Ran the script and saw that camera performs as it should.

I can't practice polar alignment on stars due to too many clouds, so I am opting for using the compass instead. Adjusted the astrotrac wedge to 16.5 degrees south. Adjusted the compass to 7 degrees eastern magnetic declination. This will have to do.

Our planned location - 2 km north of the centerline.
I'll transport the gear semi assembled, with the Borg scope in my hands in the bus so that it doesn't get shaken too much. The guests really want to help so they'll get to carry as well.

We'll have to drive 15-30 minutes south of town to a site just 2 km north of the center line. The bus leaves at 3.30AM - just 1.5 hour from now.

Just went outside and looked at the sky - it is clear! Orion and Sirius are near zenith.

Below is a NOAA IR image of the cloud situation 45 minutes ago. At that time it was cloudy here.
NOAA IR enhanced image 14.11.2012, 1:09AEST.
Red cross shows our location


Eclipse 2012 - arrival in Port Douglas!

We have just arrived in Port Douglas - the place where we will (hopefully!) be seeing the total solar eclipse. We are now just 11 hours from totality and I have a million things to do, mostly dealing with setting up the hardware. We have been in the outback and rain forest for three days out of contact with  the world.

I brought the telescopes along in the bus. That was a mistake. We've been bumping around on the roughest roads ever and the collimation must be way off by now; gotta check that shortly.

I'm back in a hotel room with Esben, our snoring tourleader. That doesn't matter much because I won't be sleeping tonight anyway. We have chartered a bus to pick us up here at 3.30AM to be at our designated site - south of here - at 4.00AM. I'll do a last minute weather satellite check and if there are clear indications that the chances will be better up north we will drive that way instead. I hope that won't be necessary.

Check out the latest weather outlook below. Scary!!!!

Latest outlook. Why do they show those nasty icons when only 'trace amount' of rain are expected and the cloud cover is 'only' 19-31%??????