Astrophotography Help Please?

[awillybilly]

Tried astrophotography for the first time tonight and it came out ok but im not satisfied. Any ideas of how I can get the fuzzy background so that it has a black background with stars and not all the red and green pixels. Any ways to fix it in photoshop if it can't be fixed with camera settings?



Edit: Sorry I didn't include many details so here they are.
Manual mode
Bulb Exposure (one shot was about 30 minutes the other was about an hour and a half)
ISO 800
F 20
picture style: Standard
White balance: AWB
Metering Mode: Evaluative Metering
Auto Focus: One Shot
Quality: Raw + Jpeg

I shot the photos with a Canon Rebel T2i EOS 550D on a tripod and used an Rc2 remote shutter release.
Im am hoping to figure out how to get similar results to this http://cdn.iwastesomuchtime.com/112220121125184.jpg

 

[TATTRAT]

Tried astrophotography for the first time tonight and it came out ok but im not satisfied. Any ideas of how I can get the fuzzy background so that it has a black background with stars and not all the red and green pixels. Any ways to fix it in photoshop if it can't be fixed with camera settings?

With no info on what you are shooting with, what your setup is, there is very little to offer except remember this: the Earth is always rotating. It can be better to stack exposures to create ONE shot. A Wide open f-stop and long shutter speed WILL leave blur (hence, better to take a lot of shorter exposures and stack), and again, without knowing your set up, the only thing I can think of for red and green pixels are inexpensive set up, if you have purple around the edges, your sensor is too warm.

What is your set up, you give zero info on that, so it's like trying to diagnose an engine problem online when you don't tell us what kind of car, year/make/model. . .we are guessing, just like you.

 

[TheFantasticG]

What is your set up, you give zero info on that, so it's like trying to diagnose an engine problem online when you don't tell us what kind of car, year/make/model. . .we are guessing, just like you.

Agreed. I opened the thread thinking I might be able to help but there's no info to go off of. Even one image might provide some info assuming the EXIF was Intact. That being said I have to agree with the stacking technique without knowing anything about the gear setup or area.

 

[thetrue]

Is stacking a cs5/6 thing, or is it possible to do in LR4 or PSE11?

 

[Edsport]

Is stacking a cs5/6 thing, or is it possible to do in LR4 or PSE11?

You can stack with deepskystacker. It's a free program...

 

[KmH]

All the red/green pixels are known as image noise.

Image noise can be caused using a high ISO setting (amplifier noise), or by making a long exposure (thermal noise).

Image sensor size and how many megapixels (MP)an image sensor has, has a direct bearing on the sensors signal-to-noise ratio (SNR). Small image sensor packed with a lot of MP means tiny pixels. the smaller pixels are, the worse the SNR is.

Camera settings that let us use a low ISO setting are a long shutter speed and a wide lens aperture. Both let in more light.

However, part of the problem is how digital photographs work. The majority of the luminosity data is in the brightest parts of the photo.
http://www.adobe.com/digitalimag/pdfs/linear_gamma.pdf
http://wwwimages.adobe.com/www.adob...ly/prophotographer/pdfs/pscs3_renderprint.pdf

 

[Patrice]

OP, you'll have to be a bit more specific with your request. Tell us what you were trying to photograph and what equipment you wee trying it with.

Astrophotography is a pretty broad field with many different 'subjects' each requiring techniques and equipment somewhat different from each other.

Do you want to photograph wide star fields like the Milky Way for instance? Specific constellations like Orion maybe? The moon? the planets? Open clusters like the Pleiades or the Double Cluster? Planetary nebulas? Emission nebulas? Double stars? Galaxies? As you can see there are quite a few different things you could be trying to photograph. Some can be done with a wide angle lens and a tripod while others might require thousands of dollars for the appropriate imaging scope and camera as well as the mount and it's guiding system and many hours worth of imaging, bias and dark frames plus the skill to process.

You might want to go look at some of the Astronomy specific forums for more information.

 

[TCampbell]

As for "fuzzy background" you'd need to post an example. The stars are moving (more accurately, WE are moving). Getting pinpoint sharp stars requires a GREAT deal of care.

Which camera do you have? Some cameras have a "long exposure noise reduction" mode. The mode causes the camera to take two images... one is the exposure you want, the second exposure is taken with the shutter CLOSED (yes - that's not a typo). It's a "dark frame". Technically a dark frame should be completely black, but since the sensor will generate image noise in a long exposure, the dark frame should help identify the noisy pixels which can then "subtracted" from the real image.

Apart from that, most astrophotographers spend a lot of time in Photoshop adjusting the image.

In high-end astrophotography, specially "cooled" cameras are used because there is a relationship between sensor temperature and the level of noise in the image. By chilling the sensor it's possible to substantially reduce a lot of the noise.

 

[TheFantasticG]

Apart from that, most astrophotographers spend a lot of time in Photoshop adjusting the image.

In high-end astrophotography, specially "cooled" cameras are used because there is a relationship between sensor temperature and the level of noise in the image. By chilling the sensor it's possible to substantially reduce a lot of the noise.

Big time on the Photoshop part. I'm in the Houston Astronomical Society. There's quite a few astrophotographers in there. The most serious ones run a liquid cooled CCD setup on a guide track system (custom or store bought kit). I just purchased a kit for $169 that allows me to sit my DSLR on TOP of my Wife's telescope. So I can have a remote and snap away while she's using her telescope to view the starts. It's a
[h=1]Celestron Advanced C8-SGT 8"/203mm Catadioptric Telescope Kit[/h]
And I enjoy looking through it but not NEARLY like she does. The kit is her christmas present as it allows a high grade spotting scope (Celestron offers a 80mm scope) that can mount to 125mm rings. The rings are detachable and I can attach a DSLR 1/4-20 mount to the vixen dovetail. Mostly wide stuff. I couldn't put a 600mm or something on it. If that was the case I get a GEM mounting plate that has a 1/4-20 stud on it.

 

[awillybilly]

Anyone have any advice after I gave more details?

 

[shefjr]

There are two great links on this forum. I have subscribed to both so that I can reference them when I need to. One deals with stars and one deals with city scrapes. Both have a plethora of great information. Enjoy!

http://www.thephotoforum.com/forum/articles-interest/276016-shooting-night-pictures-stars-stuff.html

http://www.thephotoforum.com/forum/...naheims-ultimate-guide-night-photography.html

 

[Patrice]

Anyone have any advice after I gave more details?

You don't need to shoot at f/20. The stars are so far away that for all practical purposes they are all on the same plane at infinity, so don't worry about DOF. Go ahead a open up your aperture to the sharpest one your lens has, usually about one or two stops down from wide open.

Opening up the aperture will thus allow you to shorten the exposure time considerably thereby reducing amp noise and hot pixels due to your sensor heating up. You can help things a bit by photographing during cold evenings or some other way of cooling your camera. An ice pack taped to the back will help. The earth rotates 15 degrees every hour so in 30 minutes you will see long star trails 7 1/2 degrees long. If you want photos with pin sharp stars then you either account for the rotation (search barn door tracker for an inexpensive solution) or keep your exposures short. 2 to 3 minutes max with a wide angle lens, and even shorter with longer lenses. You can calculate it using your sensor pixel pitch and the field of view of your lens. The idea is to keep any individual small star from illuminating any more that 4 to 5 pixels square. (You are not actually photographing the star but its diffraction ring as its light goes through your objective lens so naturally very bright stars will have a much larger first diffraction ring and thus appear larger in the image.) Trial and error will help you determine where the exposure time limit is for your sensor and your tastes.

For maximum contrast limit your photography to moonless nights and clear stable atmosphere. Wait a couple of hours after sunset before starting. If the stars appear to twinkle that means the 'seeing' is not very good so go watch a movie instead.

Stars have various natural colours, with white, blue and red being the most common. Use daylight white balance.

Don't be afraid of high iso - its actually not a real detriment in astrophotography. ISO noise is more or less random and astrophotography stacking software is designed, through the stacking of many exposures, to model that out. You will need to read up on 'dark frames' and 'light (bias) frames'.

edit: post a couple of your imaging attempts so that the forum can give you some help that is more specific.

 

[TCampbell]

Here's a shot I took last september on a moonless night at the Great Lakes Star Gaze in Michigan. This is a good (although not excellent) dark sky site.


Milky Way by Tim Campbell1, on Flickr

The EXIF data is intact if you want to look at it, but I took this with a Canon 5D II (full-frame camera) using a 14mm f/2.8 lens.

The shot was is a 40 second exposure at ISO 1600 at f/2.8. The lens was manually focused (auto-focus switch was turned off.)

BUT... this shot is processed. The straight-out-of-the camera shot was fairly muddy. Guys who have been doing this a lot longer than me could probably make this look a lot richer. There was more background skyglow than I wanted, but to get a TRUELY dark sky site would require quite an expedition (there are some locations in northern Michigan with amazing dark skies.)

You want to get the exposure times reasonably short, so don't use f/20.

Use your widest angle. f/2.8 is 5 & 2/3rds stops faster. That means it'll require about 1/50th of the exposure time. The problem with long exposures is you'll need a mount which tracks the stars and is aligned to Earth's polar axis otherwise you're going to get star trails. When I took my shot it was on a normal tripod (with a ballhead). This shot was not "tracking" the stars (even though I do have a camera mount which allows me to piggy-back the camera onto one of my telescopes.) At 14mm and a 40 second exposure I knew I wasn't going to get star trails... yet.

If you've got an 18-55mm kit lens then you could use 18mm & f/3.5. At ISO 1600 you'd need about a 1 minute exposure and can tweak from there. If you go to f/4 then you'd need to increase the exposure to about 1m20secs. And of course if you use ISO 800 for less "noise" then you'd need to double the exposure times, but at some point you'll start noticing elongated stars (starting to grow tails) from the movement of the sky during the long exposure.

The stars straight out above Earth's equator (declination 0) are moving at a speed of 15 arc-seconds per second of real time (that's not exact because the Earth spins 360º in only about 23 hours and 56 minutes... not 24 hours.) Every 4 seconds they move 1 arc-minute. In 4 minutes they move 1 degree (unless I screwed up my math, but I think that's right.) To help you imagine just how fast that is... the full moon is about 30 arc-minutes (yes, just 1/2 of a degree) from edge to edge. That means the moon will have moved in the sky by it's own width in just about 2 minutes! The stars nearer to the poles don't appear to moving as fast (Polaris, for example, hardly seems to move at all even though it's technically about 40 arc-seconds away from the true celestial pole (yes... if you could slide the moon up to the pole, you could fit the moon between Polaris and the true celestial north pole and still have a tiny bit of room to spare.)

Update: Ok, apparently Flickr captures the EXIF and makes it available on their site, but strips it from the image imbedded here (the image I uploaded to Flickr had the EXIF in it.) You can view the EXIF by viewing it at Flick are picking "Actions" -> "View EXIF info".