Astrophotography m4/3 - what have you got? - Page 3

jloden · #21 September 7th, 2012, 12:54 PM

Dang... it looks like Deep Sky Stacker is Windows only and I run a Mac. I might be able to use a VM to run it, though it'll complicate the workflow.

I notice neither DSS nor Nebulosity lists Panasonic RAW support; will I still be able to use RAW photos or do you need to use JPG then for stacking?

RobWatson · #22 September 7th, 2012, 01:03 PM

Quote:

Originally Posted by jloden

Some nice shots here... I tried doing star shots the other night and had a heck of a hard time getting it focused and/or getting anything interesting other than a black sky with dots. If anyone would like to share a basic primer on astrophotography I for one would be interested.

I'm not sure what you all mean about stacking exposures - I gather it's using an app to combine long exposure photos together, but I'm not familiar with the process and/or how it would differ from something like HDR (which I know from experience didn't work well with moon photos).

And Chris - are those moon shots through a scope? They look great, probably better than my very best shots with the 100-300 on a tripod so far

IRIS is free and very powerful but can have a steep learning curve.

As mentioned stacking is like HDR in that multiple images are combined to reduce noise. Does not raise blacks but lowers noise - if there is no signal captured then stacking a bunch of zeros does not create anything. With light pollution your 'blacks' will likely be quite high so dim objects will be hidden in the noise. Unlike HDR usually a single exposure time is used but many frames are combined.

I say combined instead of averaged as there are better ways to combine the frames than averaging but the basic concept is to add or average the frames. But adding or averaging the random noise is reduced. Using a method like median combining also rejects transient noise or errors (like planes, etc).

Frame stacking works best for fairly dim objects and typically not at all for bright objects (like mooon, planets, etc). Combining images can be done for bright objects but it is a specialized approach. The main problem with bright objects is atmospheric instability and high magnification which (if not too bad) creates localized (partial FOV) ares of 'distortion' sort of like heat waves. The trick is to use short exposures to 'freeze' the turbulence or capture that fleeting lucky moment of stability.

RobWatson · #23 September 7th, 2012, 01:45 PM

Quote:

Originally Posted by jloden

VM to run it, though it'll complicate the workflow.

I notice neither DSS nor Nebulosity lists Panasonic RAW support; will I still be able to use RAW photos or do you need to use JPG then for stacking?

Bootcamp probably will have a better chance.

You can do raw or jpeg ... raw is better likely as you need all the dynamic range you can get.

Djarum · #24 September 7th, 2012, 03:33 PM

Quote:

Originally Posted by RobWatson

IRIS is free and very powerful but can have a steep learning curve.

As mentioned stacking is like HDR in that multiple images are combined to reduce noise. Does not raise blacks but lowers noise - if there is no signal captured then stacking a bunch of zeros does not create anything. With light pollution your 'blacks' will likely be quite high so dim objects will be hidden in the noise. Unlike HDR usually a single exposure time is used but many frames are combined.

I say combined instead of averaged as there are better ways to combine the frames than averaging but the basic concept is to add or average the frames. But adding or averaging the random noise is reduced. Using a method like median combining also rejects transient noise or errors (like planes, etc).

Frame stacking works best for fairly dim objects and typically not at all for bright objects (like mooon, planets, etc). Combining images can be done for bright objects but it is a specialized approach. The main problem with bright objects is atmospheric instability and high magnification which (if not too bad) creates localized (partial FOV) ares of 'distortion' sort of like heat waves. The trick is to use short exposures to 'freeze' the turbulence or capture that fleeting lucky moment of stability.

Actually, high magnification moon shots are done with stacking. It helps eliminate softness due to atmospheric turbulance.

I tried using Registax one time with some photos of the moon. Worked out ok but using it is a PITA. The images have to be dead on aligned and markers have to be almost exact.

RobWatson · #25 September 7th, 2012, 04:00 PM

Quote:

Originally Posted by Djarum

Actually, high magnification moon shots are done with stacking. It helps eliminate softness due to atmospheric turbulance.

I tried using Registax one time with some photos of the moon. Worked out ok but using it is a PITA. The images have to be dead on aligned and markers have to be almost exact.

Yes, but that process rejects those images that are deemed 'too soft' so it is more complicated than a simple stack like an average, etc. Even better methods reject only portions of the frame that are deemed 'soft' so they do a kind of variable ROI and select portions of frames. That is the kind of thing I alluded to by using the word 'combination'.

I forget which issue or mag (Sky&Tel, Astronomy or CCD Astro) but Ron Dantowitz (sp?) showed a method using video in whch portions of frames were extraced and recomposited to produce remarkably sharp images of satellites and the space shuttle ... he has since been 'recruited' by NSA since he got some amazing images of super secret spy satillites - last I heard anyway.

Just simply stacking a bunch of soft images will not produce a sharp result. But taking a very large number of images and sifting through them for the best frames or even best portions of frames is a proven - if painstaking- method.

Care must be taken when stacking with planetary images since them planets rotate ...

Chris Malikoff · #26 September 7th, 2012, 04:14 PM

Quote:

Originally Posted by mowog6000

Try 60 1 minute exposures and stack with Nebulosity I got good results at f8 with the small sensor on the Orion starshoot astro camera.

Simply stacking still doesn't do the job to bring out feint detail in the periphery of nebulous objects. Stacking only reduces noise by averaging out wayward pixels. This allows you to stretch the histogram, sure, which will result in higher dynamic range and shows you more detail than by not stacking/stretching. It still doesn't expose the detail I want however. You need to go long - in addition to short. You don't and can't "add" exposure by stacking, remember that.

As Rob said - adding a bunch of zeros is still zero. You still need plenty of signal. Some of my dedicated astro CCD shots are in the vicinity of an hour - 60 full minutes on objects such as the Witch Head Nebula in Orion (off Saiph). You need REALLY dark skies to carry this off. Luckily in Australia there are stil plenty of places to go to give you this. Our local observing site, an hour from home, still affords us mag 7 skies - which is really quite dark for such close proximity to a major city - Sydney.

This is my usual setup:

Chris Malikoff · #27 September 7th, 2012, 04:19 PM

Quote:

Originally Posted by jloden

And Chris - are those moon shots through a scope? They look great, probably better than my very best shots with the 100-300 on a tripod so far

Thanks Jay. No, they're not. They are through a Nikkor ED 300mm lens from the 80's at f/4.5, stacked with a Nikon TC-300 2x teleconverter which gives a native 600mm. Add your 2x MFT multiplier and you have an effective 1,200mm lens. I then set my OM-D's IS to 1,000mm (maxed out) and hoped for the best by hand holding this combination. IBIS worked an absolute treat.

Chris Malikoff · #28 September 7th, 2012, 04:22 PM

Quote:

Originally Posted by danska

... I'm mainly interested in showing interesting foreground stuff with the Milky Way overhead so I'll be planning out some trips probably next summer to try and get some of these types of shots. I'm not sure how much of the Milky Way is really in view in the winter time.

Danska - do yourself a favour and purchase a Vixen Polarie. It is a small pocket-sized equatorial platform for your tripod. Point south (or north if you're in the northern hemisphere) and you'll be doing 2 minute wide field exposures without trailing.

http://www.vixenoptics.com/mounts/polarie.html

Djarum · #29 September 7th, 2012, 04:23 PM

Quote:

Originally Posted by RobWatson

Yes, but that process rejects those images that are deemed 'too soft' so it is more complicated than a simple stack like an average, etc. Even better methods reject only portions of the frame that are deemed 'soft' so they do a kind of variable ROI and select portions of frames. That is the kind of thing I alluded to by using the word 'combination'.

I forget which issue or mag (Sky&Tel, Astronomy or CCD Astro) but Ron Dantowitz (sp?) showed a method using video in whch portions of frames were extraced and recomposited to produce remarkably sharp images of satellites and the space shuttle ... he has since been 'recruited' by NSA since he got some amazing images of super secret spy satillites - last I heard anyway.

Just simply stacking a bunch of soft images will not produce a sharp result. But taking a very large number of images and sifting through them for the best frames or even best portions of frames is a proven - if painstaking- method.

Care must be taken when stacking with planetary images since them planets rotate ...

This was the problem I actually had with the Moon. The markers were rotating. Some programs can detect rotation and line up the markers. I think I get better results though just doing it on a night that has good seeing, but this is difficult where I live. Even though I live in the south(better seeing for planets), I live in a valley, and in the evening the cool air wants to sink, even during the summer, causing turbulence.

I've read way more than I ever wanted on cloudynights for astro imaging. I just don't have the budget or patience to do it right. The only type of astroimaging that really interests me is widefield astroimaging, but as I said previously, I have yet to get decent results because of light pollution.

RobWatson · #30 September 7th, 2012, 04:23 PM

Quote:

Originally Posted by Chris Malikoff

Thanks Jay. No, they're not. They are through a Nikkor ED 300mm lens from the 80's at f/4.5, stacked with a Nikon TC-300 2x teleconverter which gives a native 600mm. Add your 2x MFT multiplier and you have an effective 1,200mm lens. I then set my OM-D's IS to 1,000mm (maxed out) and hoped for the best by hand holding this combination. IBIS worked an absolute treat.

This IS amazing. Are you trying to give me GAS?