Camera Labs

Nice work Bob, and that's fantastic to fully remote control and focus the lens with your laptop! Keep us updated...

Hi folks,

The cloudy weather of the last few mights relented last night and I was able to try out my new 85mm lens for the first time. I'll admit straight away that there are a few technical issues with this image. The focus was very slightly off (I disabled the 40D's "LCD Auto Adjust" Live View setting which I think caused the stellar images on my laptop's screen to appear fat making it difficult to see the point of sharpest focus) and, judging by the amount of background light in the RAW images, the exposures were a little longer than they should have been. That was an issue with my local street lighting as the 120 second dark frames from the camera were totally black apart from a few stuck pixels.

Anyway, here is the result of stacking 7 images. Each exposure was 120 seconds at ISO 400 and f/1.2 with the camera mounted on top of my telescope to take advantage of the motor driven tracking. The resultant images were aligned, stacked and heavily post-processed using IRIS and a little cosmetic finishing was done using Photoshop. No airbrushing or the like was done - what you see is entirely derived from what came out of the camera. If you want to know where in the sky this object is then pop outside in the Northern Hemisphere around 11 o'clock tonight and look a little west of directly overhead (temperate latiitudes). Unfortunately the North American nebula (WiKi), seen near the left edge of the image, is too faint to be normally seen with the naked eye.



The bright star with the blue halo is Deneb and this is the 19th brightest star in the night sky. It is significantly hotter than the Sun, hence the blue colouration. Here is a map (courtesy of Wikipedia) which shows both Deneb and the North American nebula which has the designation NGC7000 and which is indicated by the larger circle. Up in my image would be roughly to the left in the map.



The combination of faster glass and stacking multiple images has certainly made a difference compared to my previous image posted earlier in this thread. I still hope to do better though. In particular, because I didn't use a "flat" frame during the image processing vignetting in the corners is far too pronounced.

Bob.

EDIT: Added comment about vignetting in the last paragraph.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Bob, that's a really nice shot regardless of it being the first with your new lens. Have you tried shooting with the aperture closed even a tad? I know it's hard to bring yourself to do it, but you may get better points, even at, say, f1.4. I found that when astro-shooting with the 85mm f1.8 - it was much better at f2.0...

It looks really promising though and I look forward to seeing what else you can get with it!

PS - have you tried an f1.2 portrait yet with the person's eye's sharp, but their ears out of focus?!

Nice shot Bob.
Stacking ? I need to brush up on my knowledge !
New lens ? F/1.2 ? at 120 seconds ? Was did done on a piggy back ?
How was it that the movement of the telescope did not cause movement or shake ?

_________________
Nikon D7000 and a bunch of Nikkors

Hi folks,

Gordon: I will certainly be trying out the lens stopped down a bit particularly if I am after really sharp stellar images. That should also help reduce the diffraction spikes which were caused by my Hutech "street light" filter being slightly smaller than the exit pupil from the lens. For my next session I want to essentially do a repeat but with with a bigger stack. I also need to understand the implications of toggling the camera's "LCD Auto Adjust" Live View setting before that last session vis-à-vis remote live view focussing. As for portraiture, the only shot so far has been of one of my dogs where his nose and eyes were out of focus but a couple of centimetres of his snout showed really sharp hairs!

Here's a couple of questions for you, Gordon. I am familiar with "dark frames" where you subtract an image taken with the camera's lens covered in order to remove objects like stuck pixels but do you use "offset frames"? See my next post for more on this. Do you use any specialist astronomical image processing software?

David: Yes, the camera was just piggy-backed on top of the telescope - the telescope lens wasn't even uncovered. The telescope's motor driven mounting automatically compensates for the Earth's rotation and for exposures of only a few minutes that is good enough to stop the stars "trailing". Because the camera was entirely controlled from the laptop the only shaking (earthquakes excepted, Gordon!) would be induced by the mirror/shutter mechanism and that damps away very quickly indeed. The stacking technique takes advantage of the fact that noise from the sensor is essentially random whilst, faint though it may be, light from faint nebulae is not. Every image you add to the stack tends to amplify the nebulae whilst tending to cancel out the random noise. If you can get the noise level down far enough then you can do that fancy image processing to remove any uniform background light (from street lights etc) and enhance the contrast. If you want more about this follow the link given the post which follows this one.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi folks,

I have mentioned several times in this thread that I use Christian Buil's IRIS software to process my astronomical images. The program is free (so this isn't a commercial ) but has quite a steep learning curve to unlock it's full potential.

For those of you who are unfamiliar with the power of techniques such as image stacking prepare to be amazed:

Before processing


After processing


These images are part of a primer on processing deep-sky DSLR images with IRIS written by Ashley Roeckelein which you can find here. The primer appears to be extremely well written and I shall certainly be studying it before my next session with IRIS. Judging from his text I still have a lot to learn.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi Bob, I'm afraid my experience with deep sky photography, dark frame subtraction and astronomical processing software is quite limited. Most of my astro work has been higher power due to lazy alignment and light polution excuses from big city locations.

But now I'm somewhere darker, I'll get round to trying them all out - although we're coming into Summer now! Unfortunately polar alignment is a lot harder down here!

Bob thats bloody awesome! great stuff.

_________________
Nikon D80+Nikkor 18-200mm VR f/3.5-5.6, Manfrotto tripod-190XDB/Head-804RC2
http://www.flickr.com/photos/hero_of_the_jay/

Those are really nice shots Bob.

_________________
Graham
Flickr

Hi folks,

Thanks for the positive feedback. I went for broke last night and took 22 x 2 minute exposures together with 10 dark frames. This time I also produced some "Flat" frames (correct for vignetting) and some "Bias" frames (correct for electronic noise. I also returned that 40D's "LCD Auto Adjust" Live View setting back to it's shipping default (enabled) which allowed me to judge the remote Live View rendition of a stellar disk much better so this time the shots are in focus. Otherwise exactly the same setup (ISO, aperture etc.) was used as for my previous post.

While the first shot hasn't been so heavily processed as the one above, and so doesn't look so pretty, the processing needed to bring out the North American nebula was still so intense that it shows some dodgy colouration at the right of the image despite my having reduced colour saturation in that area.

Both this and the next image are clickable should you wish to see the 100% originals.

Not so colourful as in the earlier post but it probably contains more information.

This image is a crop immediately around the North American nebula itself. Image processing was intense, to say the least.

Oriented this way it leaves no doubt as to how it got the name.

Judging by the dark frames (2 minute exposures at ISO400 with the lens cap on) I have in no way approached the limits of the 40D's sensor but to take advantage I would have to take the whole shooting match to a much darker spot than where I am on the outskirts of a village. Maybe one day later this year...

Bob.

EDIT: Typos corrected and I have added a comment to the effect that exactly the same setup was used as for my previous post. Thanks, Gordon.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Nice work Bob! But what aperture and ISO did you use?

Sorry, I meant to include a statement to the effect that I used exactly the same setup as with my previous post but it got left out. Apart from getting the focus right this time the big difference is that I used IRIS to process the RAW files directly rather than import the TIF files exported from Canon's DPP software - hence the need for offset frames et al. The only problem with doing it this way is that IRIS does a rotten job of producing a correctly balanced colour image. With a 16 bit resolution the IRIS files have all the data but the initial output is always heavily biased towards green (twice as many green pixels as red and blue?) so a lot of manual tweaking is needed.

Regarding ISO, if you haven't seen it check out my post here. Christian Buil seems to be of the opinion that, for low light photography in particular, ISO settings are almost irrelevant. The only caveat I would add is that the readout gain, as set by the ISO, has to be high enough to reach down into the sensor's thermal noise. For the 40D with its 14 bit A/D it seems that ISO 400 is enough.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Bob
Nice pic - and it was interesting to note that your 2nd round of capturing and processing seems to have other effects as what you mentioned, more greenish ? But i see more hot spots.
I'm surprised that processing RAW directly instead of TIFF files from Canon DPP produced less 'exciting' output, but as you mentioned - it certainly contains more details / data.

Speaking of which - how long did the post-processing took to complete this set of images ?

_________________
Nikon D7000 and a bunch of Nikkors

Most of yesterday! Mind you that was a lot to do with my learning to use IRIS in a new way and also one big mistake which took a couple of hours to figure out.

The greenish tinge is the palest shadow of the colour that IRIS produced when it converted the pre-processed stack of 22 images from RAW. On the processing side that is, in my opinion, the biggest annoyance with the program. You are left to try and point out to IRIS an area which should be white and that is less easy (for me at least) in these images than one would think.

I find it quite difficult to differentiate between hot spots and faint stars. It's easy when you look at the dark frames, of course, and these are pretty clean. Despite the heavy processing needed to emphasise the nebulosity I am quite pleased with the level of noise there. For instance, have a pixel peep at the full size download of the cropped image (here). I used some noise reduction and sharpening in Photoshop to clean up this image but given that the nebulosity is being plucked from the bottom of the pixel wells I don't think the result is too bad.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Oh I don't doubt you, Bob, or your abilities.

I was asking about the time as I wanted to be prepared of what is needed should I take up this area of photography. Way I looked at it - plenty of time.
Having taken a pixel peep I understand what you mean by identifying a white area ! Because I could not also. !

_________________
Nikon D7000 and a bunch of Nikkors