I see that photographers are photo-stacking, making several exposures
of the stars.
What I would like to know, if you make several COPIES of ONE exposure,
then carefully align them, wouldn't that be the same as several exposures?
Just a thought. Thanks.
One of the reasons astrophotographers stack images is to remove noise by averaging. Multi copies of one image would preclude this.
Manny Jay wrote:
I see that photographers are photo-stacking, making several exposures
of the stars.
What I would like to know, if you make several COPIES of ONE exposure,
then carefully align them, wouldn't that be the same as several exposures?
Just a thought. Thanks.
Is this in reference to star trail images?
I have not done this, so... take this with plenty of salt.
But it seems to me that stacking will help in two ways. One is to increase the brightness of the stars in pictures that were taken at short intervals to minimize star trails. And stacking also cuts back on sensor noise. Even if you photograph from a mount that tracks the stars, so there are no trails to worry about, stacking is still useful for reducing noise. The stars are 'fixed' in place, once the pictures are aligned, but the bright dots of noise will differ from picture to picture. So blending the pictures in a stack cuts back on the noise.
You don't need special astrophotography software to do it. But it does make it easier. See
https://astrobackyard.com/tutorials/stack-exposures/ for example.
Manny Jay wrote:
I see that photographers are photo-stacking, making several exposures
of the stars.
What I would like to know, if you make several COPIES of ONE exposure,
then carefully align them, wouldn't that be the same as several exposures?
Just a thought. Thanks.
The normal reason is to deal with random noise, using the same image fails 100% with that.
It could help in star trails, but only if the image has no foreground interest (which basically ruins the effect from my point of view).
We used to stack 35mm film negatives for the same reason, to get rid of noise--imperfect recording of grains on the film, which was random, just like a digital detector. Plus with film there was reciprocity failure--all film sensitivity (speed) rolls off rather sharply after a certain time, with the knee of the curve several seconds to a few minutes of exposure. First we would hyperprocess the film before shooting--evacuate it to dry it and remove oxygen, then backfill with forming gas (welding supply house), which was 8% hydrogen and ~92% N2, and hold in this atmosphere for 50 hours at, I think 50 degrees C??? Amateur astrophotographers invented this process, and professionals adopted it, so it became standard. I still have my small aluminum squat cylinder with a removable O-ring sealed lid and a hose port which could be attached to a hand pump to evacuate it. We'd pump it out as low a pressure as we could get and backfill with gas and repump 4-5??? times. There were commercially available tank/hose/pumps available. The tank also had a heating coil (maybe we wrapped a strap around it) to heat it when done pumping and backfilling. There was a temp regulator to maintain the temp. I would hyperprocess my film 50 hours + the time too drive to the desert dark site plus setup time, then open the can and load the film when it got dark. Of course after a 2 hour drive plus 4-6 hours setup time it was cooled down and room temp.
Verryl
Verryl
Noise is generally random, information (stars, etc.) generally are not. Noise is reduced as square root of # of pics. Also, astro software can often also select the best images to scan, so those with poor atmospheric effects are left out. Often video cameras are used, at least for planets, moon, etc., take 1000's of frames, keep best X%, good for removing changing atmospherics. There are also other methods than noise reduction used, some of which could be applied to any image I would think. For example, thefe can be hot pixels in your camera, a blank image can be taken to find them and make adjustments.
As stated before by others, astrophotography can be easy or as involved as any photography or maybe more so, dark calibration files, light response calibration, filter wheels with 4 to 8 filters, cooled cameras, guiding, etc. Except for static scences, much would not be applicable, even planets like Jupiter rotate fast enough to cause time limits on how images can be stacked.be
There are many reasonably priced books on astrophotography if interested or someone wants to explore for ideas.
There's probably close to 25 tutorials on You Tube regarding astrophotography. Some of them deal with this topic.
It would happen with any astrophotography. You can take a long exposure use the rule of 500, however that will not give you the same details that you would get from a long shoot taking multiple shots and stacking to get much more detail. It also allows for better detection of noise and reducing the effect in the final process
DirtFarmer
Loc: Escaped from the NYC area, back to MA
The camera sensor is basically a counting device. It counts the number of photons that fall onto a given point during the exposure.
Noise in a counting system which counts random events is proportional to the square root of the number counted (an approximation which starts to fall apart for numbers less than about 10).
If you use multiple exposures for a stack, each exposure contains a count at a given point that is randomly distributed about some mean. So you have several random numbers that you are adding together. Also, noise in the system is also a random number. If you have 4 exposures, the signal to noise ratio will be reduced by a factor of two. For 9 exposures you get a factor of 3 reduction, etc.
If you use a single exposure and stack copies, you do not have the random distribution. In effect, you are just multiplying the count by the number of copies. When you look at the signal to noise ratio, the number of exposures cancels out and you have the same signal to noise ratio that you would have had with one exposure.
bwana
Loc: Bergen, Alberta, Canada
Manny Jay wrote:
I see that photographers are photo-stacking, making several exposures
of the stars.
What I would like to know, if you make several COPIES of ONE exposure,
then carefully align them, wouldn't that be the same as several exposures?
Just a thought. Thanks.
The simple answer is, 'No'.
bwa
Manny Jay wrote:
I see that photographers are photo-stacking, making several exposures
of the stars.
What I would like to know, if you make several COPIES of ONE exposure,
then carefully align them, wouldn't that be the same as several exposures?
Just a thought. Thanks.
No, using the same multiples of the same photo will not reduce the noise. The basic assumption is that noise is randomly scattered across the photo and will be different from one photo to the next. By using different photos, the noise is in different locations, So when stacking, the noisy pixel on one photo will be "swamped" by the other photos that have the "correct" value at that pixel. Stacking the same photo will not reduce the noise, just keep it unchanged.
Cambridge in Colour has a nice explanation of stacking and noise reduction here:
https://www.cambridgeincolour.com/tutorials/image-averaging-noise.htm
An analogy might be imagine taking a still photo of any object. There is a fly moving around on the object, so you take 100 photos, each with fly in different place. If you now add all the data in the photos together and then divide by 100, the image data from the fly fades out compared to the other information. True noise is like that fly.
The closest effect to a fly I have seen is with a telescope which stacks on the go. I had a plane or meteor go thru early on, as more images were taken, the streak slowly faded out completely. At least some stacking software also can select only the best X% of the images to stack.
If you want to reply, then
register here. Registration is free and your account is created instantly, so you can post right away.