Short versus Long exposures

I feel that I am missing something.

As I understand it, in order to capture images of faint fuzzies, you would
need to be using a camera capable of long exposures. This is because the
shutter has to be open a long time for enough light to hit the film/ccd to
make an image.
What I dont understand is why you cant simply stack hundreds of short
exposures to build up enough light. There must be _some_ light in them?j

I seem to remember someone saying something about noise but I dont
understand it. Is there more noise in 25 frames of 1/25 of a second than a
single frame of 1 second? If so why? Or is it something else entirely.

Stuart

"Stuart M" wrote:
I feel that I am missing something.

As I understand it, in order to capture images of faint fuzzies, you would
need to be using a camera capable of long exposures. This is because the
shutter has to be open a long time for enough light to hit the film/ccd to
make an image.
What I dont understand is why you cant simply stack hundreds of short
exposures to build up enough light. There must be _some_ light in them?j

There is a minimum threshold for what you can detect. This varies with
the type of film or CCD. With all but the most exotic kit single
photons don't register - you need several to hit the CCD pixel for the
0 to become a 1 (then several more to make it a 2 etc.). The situation
with film is analogous - you need several photons to make the chemical
reaction do its magic.

If your exposure isn't long enough then very dim objects won't have a
chance to register at all, so it doesn't matter how many images you
stack as the dim object won't have registered on any of them.

Till we can all buy CCDs (or film) that can register individual
photons there will be no substitute for a long exposure when you're
trying to image dim objects.


Tim
--
Love is a travelator.

In message , Tim Auton
tim.auton@uton.[groupSexWithoutTheY] writes
"Stuart M" wrote:
I feel that I am missing something.

As I understand it, in order to capture images of faint fuzzies, you would
need to be using a camera capable of long exposures. This is because the
shutter has to be open a long time for enough light to hit the film/ccd to
make an image.
What I dont understand is why you cant simply stack hundreds of short
exposures to build up enough light. There must be _some_ light in them?j

There is a minimum threshold for what you can detect. This varies with
the type of film or CCD. With all but the most exotic kit single
photons don't register - you need several to hit the CCD pixel for the
0 to become a 1 (then several more to make it a 2 etc.). The situation
with film is analogous - you need several photons to make the chemical
reaction do its magic.

If your exposure isn't long enough then very dim objects won't have a
chance to register at all, so it doesn't matter how many images you
stack as the dim object won't have registered on any of them.

Till we can all buy CCDs (or film) that can register individual
photons there will be no substitute for a long exposure when you're
trying to image dim objects.

Actually, people are doing exactly what Stuart suggests - stacking lots
of short exposures. The big problem isn't lack of sensitivity but noise
in the image, but when you stack lots of images the noise averages out.
In fact, this is becoming a standard technique with CCDs, because it
avoids the pain of something ruining your long exposure - if you have a
problem with one of ten short exposures, you discard it. I hasten to add
that I'm not yet doing this - I'm strictly a beginner and the Moon's my
favourite target - but apparently other people are finding it works.
--
Save the Hubble Space Telescope!
Remove spam and invalid from address to reply.

Jonathan Silverlight
wrote:
In message , Tim Auton
tim.auton@uton.[groupSexWithoutTheY] writes
"Stuart M" wrote:
[snip]
What I dont understand is why you cant simply stack hundreds of short
exposures to build up enough light. There must be _some_ light in them?j

There is a minimum threshold for what you can detect.
[snip]
Till we can all buy CCDs (or film) that can register individual
photons there will be no substitute for a long exposure when you're
trying to image dim objects.

Actually, people are doing exactly what Stuart suggests - stacking lots
of short exposures.

I know, I'm one of them!

The big problem isn't lack of sensitivity but noise
in the image, but when you stack lots of images the noise averages out.

Both are problems. Stacking helps with noise (and other things), but
not with sensitivity. The OP did in effect ask what advantage long
exposures have over stacked short exposures - so I told him ;-)


Tim
--
Love is a travelator.

"Tim Auton" tim.auton@uton.[groupSexWithoutTheY] wrote in message
...
Jonathan Silverlight
wrote:
In message , Tim Auton
tim.auton@uton.[groupSexWithoutTheY] writes
"Stuart M" wrote:
[snip]
What I dont understand is why you cant simply stack hundreds of short
exposures to build up enough light. There must be _some_ light in
them?j

There is a minimum threshold for what you can detect.
[snip]
Till we can all buy CCDs (or film) that can register individual
photons there will be no substitute for a long exposure when you're
trying to image dim objects.

Actually, people are doing exactly what Stuart suggests - stacking lots
of short exposures.

I know, I'm one of them!

The big problem isn't lack of sensitivity but noise
in the image, but when you stack lots of images the noise averages out.

Both are problems. Stacking helps with noise (and other things), but
not with sensitivity. The OP did in effect ask what advantage long
exposures have over stacked short exposures - so I told him ;-)



Actually Tim, the signal (ie number of photons detecteded) is identical in
one long exposure or a stack of many short ones. Noise is always the factor
which limits sensitivity in imaging, it just comes from different sources,
some of which are averaged down by stacking and others are not. It is this
which limits the ultimate sensitivity of stacked images.

Robin

Tim Auton wrote:
Jonathan Silverlight
wrote:
In message , Tim Auton
tim.auton@uton.[groupSexWithoutTheY] writes
"Stuart M" wrote:
[snip]
What I dont understand is why you cant simply stack hundreds of
short exposures to build up enough light. There must be _some_
light in them?j

There is a minimum threshold for what you can detect.
[snip]
Till we can all buy CCDs (or film) that can register individual
photons there will be no substitute for a long exposure when you're
trying to image dim objects.

Actually, people are doing exactly what Stuart suggests - stacking
lots of short exposures.

I know, I'm one of them!

The big problem isn't lack of sensitivity but noise
in the image, but when you stack lots of images the noise averages
out.

Both are problems. Stacking helps with noise (and other things), but
not with sensitivity. The OP did in effect ask what advantage long
exposures have over stacked short exposures - so I told him ;-)

I think you are using a different, though not incorrect, definition of
sensitivity. The practical points about Stuart's question concern
how a faint fuzzy may be photographed rather than the quantum
efficiency of the detector.


--
Graham W http://www.gcw.org.uk/ PGM-FI page updated, Graphics Tutorial
WIMBORNE http://www.wessex-astro-society.freeserve.co.uk/ Wessex
Dorset UK Astro Society's Web pages, Info, Meeting Dates, Sites & Maps
Change 'news' to 'sewn' in my Reply address to avoid my spam filter.

"Stuart M" wrote in message
...


I had always thought that this "noise" was generated by the electrical
field
in the camera but then it wouldnt happen with film. What is this noise and
why is it not worse in long exposures?



Sorry Stuart,

I was specifically talking about CCD imaging. I suspect there may be
equivalent processes in chemical imaging but I am not knowledgeable in that
area

There are many sources of noise in CCD imaging. Two important ones which
apply here are thermal noise generated inthe CCD chip. This builds up with
increased exposure time, but because it is random, it builds up slower than
the non random image so longer exposures or a stack of shorter exposures
gives a lower signal/noise and so greater sensitivity. The other noise,
which limits the capability of multiple stacking, is independent of exposure
time and is generated in the electronics each time a frame is downloaded.
This of course only occurs once in a single long exposure but adds up for
stacked images. It is for this reason that stacks can never be as good as an
equivalent single long exposure, and ultimately limits the maximum
sensitivity of stacked short exposures.

Robin

"Robin Leadbeater" wrote in message
...
...
There are many sources of noise in CCD imaging. Two important ones which
apply here are thermal noise generated inthe CCD chip. This builds up
with
increased exposure time, but because it is random, it builds up slower
than
the non random image so longer exposures or a stack of shorter exposures
gives a lower signal/noise and so greater sensitivity.

Err.. that should read lower noise to signal (ratio) ... but more commonly
expressed as higher signal to noise (ratio).


--
Graham W http://www.gcw.org.uk/ PGM-FI page updated, Graphics Tutorial
WIMBORNE http://www.wessex-astro-society.freeserve.co.uk/ Wessex
Dorset UK Astro Society's Web pages, Info, Meeting Dates, Sites & Maps
Change 'news' to 'sewn' in my Reply address to avoid my spam filter.

Wasn't it Stuart M who wrote:

I had always thought that this "noise" was generated by the electrical field
in the camera but then it wouldnt happen with film. What is this noise and
why is it not worse in long exposures?

One of the things that's going on with photographic film is that there's
a background of high energy particles whizzing around you all the time.
Some of these are powerful enough to penetrate the film canister before
or after you have exposed the image.

If you're taking holiday snaps of bright beach scenes, then a tiny
amount of fogging from the airport X-Ray machine will be unnoticeable.
If you're going to try stacking large numbers of photographic plates
that contain very faint images, then minute amounts of noise from things
like cosmic rays and granite rocks could build up.

CCD cameras will also pick up such energetic particles, but any
accumulated values are reset to zero when you start an exposure.
Accumulated data from alpha particles that have hit a piece of film
don't get wiped out when you open the shutter, so the effect doesn't
depend on exposure time.

Another source of film noise is dust. Unless you handle your camera and
film in an ultraclean environment, you can get fine dust particles on
the film. The effect of the dust does not depend on the exposure time.

--
Mike Williams
Gentleman of Leisure

"Robin Leadbeater" wrote in message
...

"Stuart M" wrote in message
...


I had always thought that this "noise" was generated by the electrical
field
in the camera but then it wouldnt happen with film. What is this noise
and
why is it not worse in long exposures?



Sorry Stuart,

I was specifically talking about CCD imaging. I suspect there may be
equivalent processes in chemical imaging but I am not knowledgeable in
that
area

There are many sources of noise in CCD imaging. Two important ones which
apply here are thermal noise generated inthe CCD chip. This builds up with
increased exposure time, but because it is random, it builds up slower
than
the non random image so longer exposures or a stack of shorter exposures
gives a lower signal/noise and so greater sensitivity. The other noise,
which limits the capability of multiple stacking, is independent of
exposure
time and is generated in the electronics each time a frame is downloaded.
This of course only occurs once in a single long exposure but adds up for
stacked images. It is for this reason that stacks can never be as good as
an
equivalent single long exposure, and ultimately limits the maximum
sensitivity of stacked short exposures.

Robin


Thank you Robin. That was very informative.