Oknoder Loc: Western North Dakota

First off let me reiterate that I am, what I consider to be a newbie to astroimaging. This past April marks my first year dedicating time, money, energy and most of all sanity to this new emerging hobby of mine. While I may not be an expert I do pride myself in my research abilities when I set my mind to something I want to learn, so I have read almost everything I can get my hands on, I will even watch countless videos, but most importantly I play with different programs and their countless options to see what effect they have on some of my images.

Darks:

I will start with Darks, since they are by far the most important part of the noise removal process. These are usually advised to be taken at the same time as the Light frames of your astroimage. The reason for this is primarily the temperature of the sensor at the time of imaging. With most higher end astrocams they utilize TEC or thermo electrically cooled systems to help reduce the noise. I read somewhere and am of the belief that for every increase of 7 degrees Celsius the noise inherent in your sensor is doubled.

It is necessary to calibrate each of your light images with 20-30 Darks of the same exposure time and the same ISO. These also should be taken around the same temperature as the Light image was captured at. This helps Astro software determine what is signal and what is noise. In theory this is exactly what long exposure noise reduction does when you take an image and then have to wait the same amount of time of the exposure before you are able to take another one. Personally I have read not to take images with noise reduction turned on, and honestly since I use Nikon, I do not hold too much faith in their ability to put out a decent piece of software. Their cameras are top notch but their programs leave a bit to be desired, IMHO.

I normally take my darks while I am setting up everything or while I am tearing it down depending on how much time I have determines the amount of frames I capture. I usually try for 50-100 but get bored and quit around 10-20, lol. Based on calculations by John Smith on his website, http://www.hiddenloft.com/ (an excellent resource for CCD imagers), for a typical camera/exposure combination, 3-4 darks gives approximately a 10% contribution from dark noise. 6-8 darks reduces this amount to 5%, and to obtain a 1% contribution, 20-30 darks must be used.

Bias:

These are by far the easiest to acquire, images captured with the shortest shutter speed with the lowest ISO. As with the darks these need to be taken when the lens, scope or camera body are covered from any light source. Bias frames are used to eliminate the read noise generated by the camera simply reading the electronic noise produced by the shutter and noise produced offloading the electron packets from the well of each pixel the camera must read to interpolate into an image. The experts say to take around 50 or so I usually shoot about 100 or so. These do not need to be taken at every imaging session like the darks and flats should be. Once or twice a year is more than enough as the read noise rarely changes too much over the course of a camera's life span. Software, such as DSS will average these images together into a master which is then subtracted from all of your light frames.

Flats:

Saving the best for last, Flats are used to record the uneven illumination due to the vignetting that most optic systems have, and due to the obstructions of the light caused by dust particles (dust donuts) either on the glass/mirrors of your optics, filters or even the sensor itself. With daytime photography these are seldom noticed but when stacking 100 images these dustbunnies stick out like a sore thumb. There are many different ways to take flats, some use EL(Electo-luminecent) panels, which can be a bit pricey if buying through an astro dealer or can be made in a DIY fashion. Others use specially constructed light boxes to achieve an even illuminated field. Personally I use a poorman's approach, I take a white T-shirt stretched over the front of the scope beginning at the crack of dawn. There are programs that will acquire these images for you by determining the average illumination and shooting 10-20 images. The key is to not move your camera's rotation until you have captured your flats. I repeatedly forget to do this, as when I get my target on sensor, I will compose the image, many times using a rotational motor on my focuser to change the orientation of the subject in comparison with the background sky, sometimes to eliminate a very bright star or to include another object in the close vicinity. As an example when imaging the Horsehead nebula many will also try to include the Flame nebula as well, depending on your focal length this can be a bit tricky. So when I move my camera's rotation without taking the flats first all the vignetting and dust donuts will be in the wrong position and tend to destroy data instead of helping preserve it and help eliminate the gradients that plague most astroimagers.

Hopefully I did not confound the issue even more but if I did here is an excellent tutorial from Don Reed about the process from beginning to end.

http://www.pbase.com/dtreed/how_its_done


Clear Skies, guys
Matthew

Europa Loc: West Hills, CA

Great post and a very good reminder on how taking the few extra "minutes" can provide a huge impact.

SonnyE Loc: Communist California, USA

I took the time to shoot dark frames with PHD last night for the guide camera.
Then promptly kicked my butt for not doing it before... :hunf:

Made the camera's view much more defined, and the guide stars stand out. (I was back on Saturn again.)
(I'm like spit on stucco. I get stuck and just hang on. :twisted: )

Albuqshutterbug Loc: Albuquerque NM

So now you're a juicy "Klingon"?
Great image their Sonny.

:shock:

Europa Loc: West Hills, CA

:lol:

CraigFair Loc: Santa Maria, CA.

A lot to remember Matthew but very informative.
Thank you.
Craig

Straight Shooter Loc: Newfoundland, Canada

A very lucid explanation of the process. I've bookmarked it.
Thanks for that!